Trajectory-Guided image-to-video (I2V) generation aims to synthesize videos that adhere to user-specified motion instructions. Existing methods typically rely on computationally expensive fine-tuning on scarce annotated datasets. Although some zero-shot methods attempt to trajectory control in the latent space, they may yield unrealistic motion by neglecting 3D perspective and creating a misalignment between the manipulated latents and the network's noise predictions. To address these challenges, we introduce Zo3T, a novel zero-shot test-time-training framework for trajectory-guided generation with three core innovations: First, we incorporate a 3D-Aware Kinematic Projection, leveraging inferring scene depth to derive perspective-correct affine transformations for target regions. Second, we introduce Trajectory-Guided Test-Time LoRA, a mechanism that dynamically injects and optimizes ephemeral LoRA adapters into the denoising network alongside the latent state. Driven by a regional feature consistency loss, this co-adaptation effectively enforces motion constraints while allowing the pre-trained model to locally adapt its internal representations to the manipulated latent, thereby ensuring generative fidelity and on-manifold adherence. Finally, we develop Guidance Field Rectification, which refines the denoising evolutionary path by optimizing the conditional guidance field through a one-step lookahead strategy, ensuring efficient generative progression towards the target trajectory. Zo3T significantly enhances 3D realism and motion accuracy in trajectory-controlled I2V generation, demonstrating superior performance over existing training-based and zero-shot approaches.

Recent advancements in Large Language Models (LLMs) have shown that it is promising to utilize Process Reward Models (PRMs) as verifiers to enhance the performance of LLMs. However, current PRMs face three key challenges: (1) limited process supervision and generalization capabilities, (2) dependence on scalar value prediction without leveraging the generative abilities of LLMs, and (3) inability to scale the test-time compute of PRMs. In this work, we introduce GenPRM, a generative process reward model that performs explicit Chain-of-Thought (CoT) reasoning with code verification before providing judgment for each reasoning step. To obtain high-quality process supervision labels and rationale data, we propose Relative Progress Estimation (RPE) and a rationale synthesis framework that incorporates code verification. Experimental results on ProcessBench and several mathematical reasoning tasks show that GenPRM significantly outperforms prior PRMs with only 23K training data from MATH dataset. Through test-time scaling, a 1.5B GenPRM outperforms GPT-4o, and a 7B GenPRM surpasses Qwen2.5-Math-PRM-72B on ProcessBench. Additionally, GenPRM demonstrates strong abilities to serve as a critic model for policy model refinement. This work establishes a new paradigm for process supervision that bridges the gap between PRMs and critic models in LLMs.

The rapid proliferation of short videos on social media platforms presents unique challenges and opportunities for recommendation systems. Users exhibit diverse preferences, and the responses resulting from different strategies often conflict with one another, potentially exhibiting inverse correlations between metrics such as watch time and video view counts. Existing uplift models face limitations in handling the heterogeneous multi-treatment scenarios of short-video recommendations, often failing to effectively capture both the synergistic and individual causal effects of different strategies. Furthermore, traditional fixed-weight approaches for balancing these responses lack personalization and can result in biased decision-making. To address these issues, we propose a novel Heterogeneous Multi-treatment Uplift Modeling (HMUM) framework for trade-off optimization in short-video recommendations. HMUM comprises an Offline Hybrid Uplift Modeling (HUM) module, which captures the synergistic and individual effects of multiple strategies, and an Online Dynamic Decision-Making (DDM) module, which estimates the weights of different user responses in real-time for personalized decision-making. Evaluated on two public datasets, an industrial dataset, and online A/B experiments on the Kuaishou platform, our model demonstrated superior offline performance and significant improvements in key metrics. It is now fully deployed on the platform, benefiting hundreds of millions of users.

Sequential recommendation models have been widely adopted for modeling user behavior. Existing approaches typically construct user interaction sequences by sorting items according to timestamps and then model user preferences from historical behaviors. While effective, such a process only considers the order of temporal information but overlooks the actual time spans between interactions, resulting in a coarse representation of users' temporal dynamics and limiting the model's ability to capture long-term and short-term interest evolution. To address this limitation, we propose RoTE, a novel multi-level temporal embedding module that explicitly models time span information in sequential recommendation. RoTE decomposes each interaction timestamp into multiple temporal granularities, ranging from coarse to fine, and incorporates the resulting temporal representations into item embeddings. This design enables models to capture heterogeneous temporal patterns and better perceive temporal distances among user interactions during sequence modeling. RoTE is a lightweight, plug-and-play module that can be seamlessly integrated into existing Transformer-based sequential recommendation models without modifying their backbone architectures. We apply RoTE to several representative models and conduct extensive experiments on three public benchmarks. Experimental results demonstrate that RoTE consistently enhances the corresponding backbone models, achieving up to a 20.11% improvement in NDCG@5, which confirms the effectiveness and generality of the proposed approach. Our code is available at https://github.com/XiaoLongtaoo/RoTE.

This paper concentrates on Multi-modal Referring Video Segmentation task, where a well optimized model is able to recognize and segment the target objects referred by the given guidance signals, e.g., language description. Early approaches model this task as a sequence prediction problem. The lack of a global view of video content leads to difficulties in effectively utilizing inter-frame relationships. Some recent works propose to perform temporal modeling with vanilla attention mechanism. However, the condensed visual representation tends to be messy about target information due to occlusion or motion blur. Unlimited non-local operation would spread such noise to all the sequences and interfere with the extraction of global representations. To address the above issue, we present Semantic-assisted Object Cluster network (SOC) and the improved SOC++ in this paper. Our method unifies temporally selective interaction and cross-modal alignment to achieve video-level understanding. In SOC++, a proxy-assisted multi-modal fusion module is introduced to perform preliminary bidirectional activation. Then a semantic integration module with progressive frame-to-video structure facilitates joint space learning across modalities and time steps. Considering that potential noisy visual embeddings would impair the overall representation of target objects in unconstrained inter-frame interactions, we propose to perform tendentious video aggregation through emphasizing the indicative role of the informative frames with lower entropy in this part. A multi-modal query contrastive supervision is also utilized to help construct well-aligned joint space at the video level. Moreover, to integrate the advantage of high-level video information and the low-level details of each frame, we introduce a dynamic query fusion module that performs joint updating of these embeddings. We conduct extensive experiments on popular referring video segmentation benchmarks, and our method outperforms state-of-the-art competitors on all benchmarks by a remarkable margin. Besides, the emphasis on temporal coherence enhances the segmentation stability and adaptability of our method in processing text expressions with temporal variations..

Recently, deep neural networks have greatly advanced undersampled Magnetic Resonance Image (MRI) reconstruction, wherein most studies follow the one-anatomy-one-network fashion, i.e., each expert network is trained and evaluated for a specific anatomy. Apart from inefficiency in training multiple independent models, such convention ignores the shared de-aliasing knowledge across various anatomies which can benefit each other. To explore the shared knowledge, one naive way is to combine all the data from various anatomies to train an all-round network. Unfortunately, despite the existence of the shared de-aliasing knowledge, we reveal that the exclusive knowledge across different anatomies can deteriorate specific reconstruction targets, yielding overall performance degradation. Observing this, in this study, we present a novel deep MRI reconstruction framework with both anatomy-shared and anatomy-specific parameterized learners, aiming to "seek common ground while reserving differences" across different anatomies. Particularly, the primary anatomy-shared learners are exposed to different anatomies to model rich shared de-aliasing knowledge, while the efficient anatomy-specific learners are trained with their target anatomy for exclusive knowledge. Four different implementations of anatomy-specific learners are presented and explored on the top of our framework in two MRI reconstruction networks. Comprehensive experiments on brain, knee and cardiac MRI datasets demonstrate that three of these learners are able to enhance reconstruction performance via multiple anatomy collaborative learning. Extensive studies show that our strategy can also benefit multiple pulse sequence MRI reconstruction by integrating sequence-specific learners.

Recent years have witnessed significant advancements in text-guided style transfer, primarily attributed to innovations in diffusion models. These models excel in conditional guidance, utilizing text or images to direct the sampling process. Traditional style transfer focuses on low-level visual features, such as brushstroke textures and color distributions, and appears more like applying an artistic filter to an image. Artistic attribute transfer, however, transcends the limitations of traditional style transfer by achieving the transfer of visual concepts from color and brushstrokes to high level aesthetic attributes such as composition, pose, and key semantic elements, resulting in more natural outcomes. Therefore, we propose an innovative text-to-image artistic attribute transfer framework named ArtCrafter. Specifically, we introduce an attention-based style extraction module, meticulously engineered to capture the subtle artistic attribute elements within an image. This module features a multi-layer architecture that leverages the capabilities of perceiver attention mechanisms to integrate fine-grained information. Additionally, we present a novel text-image aligning augmentation component that adeptly balances control over both modalities, enabling the model to efficiently map image and text embeddings into a shared feature space. We achieve this through attention operations that enable smooth information flow between modalities. Lastly, we incorporate an explicit modulation that seamlessly blends multimodal enhanced embeddings with original embeddings through an embedding reframing design, empowering the model to generate diverse outputs. Extensive experiments demonstrate that ArtCrafter yields impressive results in visual stylization, exhibiting exceptional levels of artistic attribute intensity, controllability, and diversity.

This paper introduces MultiBooth, a method that generates images from texts containing various concepts from users.Despite diffusion models bringing significant advancements for customized text-to-image generation, existing methods often struggle with multi-concept scenarios due to low concept fidelity and high inference cost. MultiBooth addresses these issues by dividing the multi-concept generation process into two phases: a single-concept learning phase and a multi-concept integration phase. During the single-concept learning phase, we employ a multi-modal image encoder and an efficient concept encoding technique to learn a concise and discriminative representation for each concept. In the multi-concept integration phase, we use bounding boxes to define the generation area for each concept within the cross-attention map. This method enables the creation of individual concepts within their specified regions, thereby facilitating the formation of multi-concept images. This strategy not only improves concept fidelity but also reduces additional inference cost. MultiBooth surpasses various baselines in both qualitative and quantitative evaluations, showcasing its superior performance and computational efficiency.

In the domain of computer vision, Parameter-Efficient Tuning (PET) is increasingly replacing the traditional paradigm of pre-training followed by full fine-tuning. PET is particularly favored for its effectiveness in large foundation models, as it streamlines transfer learning costs and optimizes hardware utilization. However, the current PET methods are mainly designed for single-modal optimization. While some pioneering studies have undertaken preliminary explorations, they still remain at the level of aligned encoders (e.g., CLIP) and lack exploration of misaligned encoders. These methods show sub-optimal performance with misaligned encoders, as they fail to effectively align the multimodal features during fine-tuning. In this paper, we introduce DETRIS, a parameter-efficient tuning framework designed to enhance low-rank visual feature propagation by establishing dense interconnections between each layer and all preceding layers, which enables effective cross-modal feature interaction and adaptation to misaligned encoders. We also suggest using text adapters to improve textual features. Our simple yet efficient approach greatly surpasses state-of-the-art methods with 0.9% to 1.8% backbone parameter updates, evaluated on challenging benchmarks.

The performance of offline reinforcement learning (RL) suffers from the limited size and quality of static datasets. Model-based offline RL addresses this issue by generating synthetic samples through a dynamics model to enhance overall performance. To evaluate the reliability of the generated samples, uncertainty estimation methods are often employed. However, model ensemble, the most commonly used uncertainty estimation method, is not always the best choice. In this paper, we propose a Search-based Uncertainty estimation method for Model-based Offline RL (SUMO) as an alternative. SUMO characterizes the uncertainty of synthetic samples by measuring their cross entropy against the in-distribution dataset samples, and uses an efficient search-based method for implementation. In this way, SUMO can achieve trustworthy uncertainty estimation. We integrate SUMO into several model-based offline RL algorithms including MOPO and Adapted MOReL (AMOReL), and provide theoretical analysis for them. Extensive experimental results on D4RL datasets demonstrate that SUMO can provide accurate uncertainty estimation and boost the performance of base algorithms. These indicate that SUMO could be a better uncertainty estimator for model-based offline RL when used in either reward penalty or trajectory truncation.

Recent 3D content generation pipelines commonly employ Variational Autoencoders (VAEs) to encode shapes into compact latent representations for diffusion-based generation. However, the widely adopted uniform point sampling strategy in Shape VAE training often leads to a significant loss of geometric details, limiting the quality of shape reconstruction and downstream generation tasks. We present Dora-Vae, a novel approach that enhances VAE reconstruction through our proposed sharp edge sampling strategy and a dual cross-attention mechanism. By identifying and prioritizing regions with high geometric complexity during training, our method significantly improves the preservation of fine-grained shape features. Such sampling strategy and the dual attention mechanism enable the VAE to focus on crucial geometric details that are typically missed by uniform sampling approaches. To systematically evaluate VAE reconstruction quality, we additionally propose Dora-Bench, a benchmark that quantifies shape complexity through the density of sharp edges, introducing a new metric focused on reconstruction accuracy at these salient geometric features. Extensive experiments on the Dora-Bench demonstrate that Dora-Vae achieves comparable reconstruction quality to the state-of-the-art dense XCube-Vae while requiring a latent space at least 8× smaller (1,280 vs. > 10,000 codes). Project page: https://aruichen.github.io/Dora.

Recently, text-to-motion models have opened new possibilities for creating realistic human motion with greater efficiency and flexibility. However, aligning motion generation with event-level textual descriptions presents unique challenges due to the complex relationship between textual prompts and desired motion outcomes. To address this, we introduce AToM, a framework that enhances the alignment between generated motion and text prompts by leveraging reward from GPT-4Vision. AToM comprises three main stages: Firstly, we construct a dataset MotionPreferthat pairs three types of event-level textual prompts with generated motions, which cover the integrity, temporal relationship and frequency of motion. Secondly, we design a paradigm that utilizes GPT-4Vision for detailed motion annotation, including visual data formatting, task-specific instructions and scoring rules for each sub-task. Finally, we fine-tune an existing text-to-motion model using reinforcement learning guided by this paradigm. Experimental results demonstrate that AToM significantly improves the event-level alignment quality of text-to-motion generation. Project page is available at https://atom-motion.github.io/.

Recent advances in RGBD-based category-level object pose estimation have been limited by their reliance on precise depth information, restricting their broader applicability. In response, RGB-based methods have been developed. Among these methods, geometry-guided pose regression that originated from instance-level tasks has demonstrated strong performance. However, we argue that the NOCS map is an inadequate intermediate representation for geometry-guided pose regression method, as its many-to-one correspondence with category-level pose introduces redundant instance-specific information, resulting in suboptimal results. This paper identifies the intra-class variation problem inherent in pose regression based solely on the NOCS map and proposes the Intra-class Variation-Free Consensus (IVFC) map, a novel coordinate representation generated from the category-level consensus model. By leveraging the complementary strengths of the NOCS map and the IVFC map, we introduce GIVEPose, a framework that implements Gradual Intra-class Variation Elimination for category-level object pose estimation. Extensive evaluations on both synthetic and real-world datasets demonstrate that GIVEPose significantly outperforms existing state-of-the-art RGB-based approaches, achieving substantial improvements in category-level object pose estimation. Our code is available at https://github.com/ziqin-h/GIVEPose.

Recent portrait animation methods have made significant strides in generating realistic lip synchronization. However, they often lack explicit control over head movements and facial expressions, and cannot produce videos from multiple viewpoints, resulting in less controllable and expressive animations. Moreover, text-guided portrait animation remains underexplored, despite its user-friendly nature. We present a novel two-stage text-guided framework, MVPortrait (Multi-view Vivid Portrait), to generate expressive multi-view portrait animations that faithfully capture the described motion and emotion. MVPortrait is the first to introduce FLAME as an intermediate representation, effectively embedding facial movements, expressions, and view transformations within its parameter space. In the first stage, we separately train the FLAME motion and emotion diffusion models based on text input. In the second stage, we train a multi-view video generation model conditioned on a reference portrait image and multi-view FLAME rendering sequences from the first stage. Experimental results exhibit that MVPortrait outperforms existing methods in terms of motion and emotion control, as well as view consistency. Furthermore, by leveraging FLAME as a bridge, MVPortrait becomes the first controllable portrait animation framework that is compatible with text, speech, and video as driving signals.

Multi-layer image generation is a fundamental task that enables users to isolate, select, and edit specific image layers, thereby revolutionizing interactions with generative models. In this paper, we introduce the Anonymous Region Transformer (ART), which facilitates the direct generation of variable multi-layer transparent images based on a global text prompt and an anonymous region layout. Inspired by Schema theory1, this anonymous region layout allows the generative model to autonomously determine which set of visual tokens should align with which text tokens, which is in contrast to the previously dominant semantic layout for the image generation task. In addition, the layer-wise region crop mechanism, which only selects the visual tokens belonging to each anonymous region, significantly reduces attention computation costs and enables the efficient generation of images with numerous distinct layers (e.g., 50+). When compared to the full attention approach, our method is over 12 times faster and exhibits fewer layer conflicts. Furthermore, we propose a high-quality multi-layer transparent image autoencoder that supports the direct encoding and decoding of the transparency of variable multi-layer images in a joint manner. By enabling precise control and scalable layer generation, ART establishes a new paradigm for interactive content creation.

With the explosive growth of 3D content creation, there is an increasing demand for automatically converting static 3D models into articulation-ready versions that support realistic animation. Traditional approaches rely heavily on manual annotation, which is both time-consuming and labor-intensive. Moreover, the lack of large-scale benchmarks has hindered the development of learning-based solutions. In this work, we present MagicArticulate, an effective framework that automatically transforms static 3D models into articulation-ready assets. Our key contributions are threefold. First, we introduce Articulation-XL, a large-scale benchmark containing over 33k 3D models with high-quality articulation annotations, carefully curated from Objaverse-XL. Second, we propose a novel skeleton generation method that formulates the task as a sequence modeling problem, leveraging an autoregressive transformer to naturally handle varying numbers of bones or joints within skeletons and their inherent dependencies across different 3D models. Third, we predict skinning weights using a functional diffusion process that incorporates volumetric geodesic distance priors between vertices and joints. Extensive experiments demonstrate that MagicArticulate significantly outperforms existing methods across diverse object categories, achieving high-quality articulation that enables realistic animation. Project page: https://chaoyuesong.github.io/MagicArticulate.

Traditional reinforcement learning methods for human- object interaction (HOI) rely on labor-intensive, manually designed skill rewards that do not generalize well across different interactions. We introduce SkillMimic, a unified data-driven framework that fundamentally changes how agents learn interaction skills by eliminating the need for skill-specific rewards. Our key insight is that a unified HOI imitation reward can effectively capture the essence of diverse interaction patterns from HOI datasets. This enables SkillMimic to learn a single policy that not only masters multiple interaction skills but also facilitates skill transitions, with both diversity and generalization improving as the HOI dataset grows. For evaluation, we collect and introduce two basketball datasets containing approximately 35 minutes of diverse basketball skills. Extensive experiments show that SkillMimic successfully masters a wide range of basketball skills including stylistic variations in dribbling, layup, and shooting. Moreover, these learned skills can be effectively composed by a high-level controller to accomplish complex and long-horizon tasks such as consecutive scoring, opening new possibilities for scalable and generalizable interaction skill learning. Project page: https://ingrid789.github.io/SkillMimic/

We introduce HunyuanPortrait, a diffusion-based condition control method that employs implicit representations for highly controllable and lifelike portrait animation. Given a single portrait image as an appearance reference and video clips as driving templates, HunyuanPortrait can animate the character in the reference image by the facial expression and head pose of the driving videos. In our framework, we utilize pre-trained encoders to achieve the decoupling of portrait motion information and identity in videos. To do so, implicit representation is adopted to encode motion information and is employed as control signals in the animation phase. By leveraging the power of stable video diffusion as the main building block, we carefully design adapter layers to inject control signals into the denoising unet through attention mechanisms. These bring spatial richness of details and temporal consistency. HunyuanPortrait also exhibits strong generalization performance, which can effectively disentangle appearance and motion under different image styles. Our framework outperforms existing methods, demonstrating superior temporal consistency and controllability. Our project is available at HunyuanPortrait.

Kechen Jiao, Zhirui Fang, Jiahao Liu, Bei Li, Qifan Wang, Xinyu Liu, Junhao Ruan, Zhongjian Qiao, Yifan Zhu, Yaxin Xu, Jingang Wang, Xiu Li. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Reward engineering is one of the key challenges in Reinforcement Learning (RL).Preference-based RL effectively addresses this issue by learning from human feedback.However, it is both time-consuming and expensive to collect human preference labels.In this paper, we propose a novel Vision-Language Preference learning framework, named VLP, which learns a vision-language preference model to provide feedback for embodied manipulation tasks.To achieve this, we define three types of language-conditioned preferences and construct a vision-language preference dataset, which contains versatile implicit preference orders.The model learns to extract languagerelated features, and then serves as a predictor in various downstream tasks.The policy can be learned according to the annotated labels via reward learning or direct policy optimization.Extensive empirical results on simulated embodied manipulation tasks demonstrate that our method provides accurate preferences and generalizes to unseen tasks and unseen language instructions, outperforming the baselines by a large margin and shifting the burden from continuous, per-task human annotation to one-time, per-domain data collection.

Traditional image-to-3D models often struggle with scenes containing multiple objects due to biases and occlusion complexities. To address this challenge, we present REPARO, a novel approach for compositional 3D asset generation from single images. REPARO employs a two-step process: first, it extracts individual objects from the scene and reconstructs their 3D meshes using off-the-shelf image-to-3D models; then, it optimizes the layout of these meshes through differentiable rendering techniques, ensuring coherent scene composition. By integrating optimal transport-based long-range appearance loss term and high-level semantic loss term in the differentiable rendering, REPARO can effectively recover the layout of 3D assets. The proposed method can significantly enhance object independence, detail accuracy, and overall scene coherence. Extensive evaluation of multi-object scenes demonstrates that our REPARO offers a comprehensive approach to address the complexities of multi-object 3D scene generation from single images.

Talking head synthesis is vital for virtual avatars and human-computer interaction. However, most existing methods are typically limited to accepting control from a single primary modality, restricting their practical utility. To this end, we introduce \textbf{ACTalker}, an end-to-end video diffusion framework that supports both multi-signals control and single-signal control for talking head video generation. For multiple control, we design a parallel mamba structure with multiple branches, each utilizing a separate driving signal to control specific facial regions. A gate mechanism is applied across all branches, providing flexible control over video generation. To ensure natural coordination of the controlled video both temporally and spatially, we employ the mamba structure, which enables driving signals to manipulate feature tokens across both dimensions in each branch. Additionally, we introduce a mask-drop strategy that allows each driving signal to independently control its corresponding facial region within the mamba structure, preventing control conflicts. Experimental results demonstrate that our method produces natural-looking facial videos driven by diverse signals and that the mamba layer seamlessly integrates multiple driving modalities without conflict. The project website can be found at https://harlanhong.github.io/publications/actalker/index.html.

Extracting physically plausible 3D human motion from videos is a critical task. Although existing simulation-based motion imitation methods can enhance the physical quality of daily motions estimated from monocular video capture, extending this capability to high-difficulty motions remains an open challenge. This can be attributed to some flawed motion clips in video-based motion capture results and the inherent complexity in modeling high-difficulty motions. Therefore, sensing the advantage of segmentation in localizing human body, we introduce a mask-based motion correction module (MCM) that leverages motion context and video mask to repair flawed motions, producing imitation-friendly motions; and propose a physics-based motion transfer module (PTM), which employs a pretrain and adapt approach for motion imitation, improving physical plausibility with the ability to handle in-the-wild and challenging motions. Our approach is designed as a plug-and-play module to physically refine the video motion capture results, including high-difficulty in-the-wild motions. Finally, to validate our approach, we collected a challenging in-the-wild test set to establish a benchmark, and our method has demonstrated effectiveness on both the new benchmark and existing public datasets.https://physicalmotionrestoration.github.io

Recent advances in Customized Concept Swapping (CCS) enable a text-to-image model to swap a concept in the source image with a customized target concept. However, the existing methods still face the challenges of $\textit{\textbf{inconsistency}}$ and $\textit{\textbf{inefficiency}}$. They struggle to maintain consistency in both the foreground and background during concept swapping, especially when the shape difference is large between objects. Additionally, they either require time-consuming training processes or involve redundant calculations during inference. To tackle these issues, we introduce InstantSwap, a new CCS method that aims to handle sharp shape disparity at speed. Specifically, we first extract the bbox of the object in the source image $\textit{automatically}$ based on attention map analysis and leverage the bbox to achieve both foreground and background consistency. For background consistency, we remove the gradient outside the bbox during the swapping process so that the background is free from being modified. For foreground consistency, we employ a cross-attention mechanism to inject semantic information into both source and target concepts inside the box. This helps learn semantic-enhanced representations that encourage the swapping process to focus on the foreground objects. To improve swapping speed, we avoid computing gradients at each timestep but instead calculate them periodically to reduce the number of forward passes, which improves efficiency a lot with a little sacrifice on performance. Finally, we establish a benchmark dataset to facilitate comprehensive evaluation. Extensive evaluations demonstrate the superiority and versatility of InstantSwap.

Illumination degradation image restoration (IDIR) techniques aim to improve the visibility of degraded images and mitigate the adverse effects of deteriorated illumination. Among these algorithms, diffusion-based models (DM) have shown promising performance but are often burdened by heavy computational demands and pixel misalignment issues when predicting the image-level distribution. To tackle these problems, we propose to leverage DM within a compact latent space to generate concise guidance priors and introduce a novel solution called Reti-Diff for the IDIR task. Specifically, Reti-Diff comprises two significant components: the Retinex-based latent DM (RLDM) and the Retinex-guided transformer (RGformer). RLDM is designed to acquire Retinex knowledge, extracting reflectance and illumination priors to facilitate detailed reconstruction and illumination correction. RGformer subsequently utilizes these compact priors to guide the decomposition of image features into their respective reflectance and illumination components. Following this, RGformer further enhances and consolidates these decomposed features, resulting in the production of refined images with consistent content and robustness to handle complex degradation scenarios. Extensive experiments demonstrate that Reti-Diff outperforms existing methods on three IDIR tasks, as well as downstream applications.

We explore cross-domain offline reinforcement learning (RL) where offline datasets from another domain can be accessed to facilitate policy learning. However, the underlying environments of the two datasets may have dynamics mismatches, incurring inferior performance when simply merging the data of two domains. Existing methods mitigate this issue by training domain classifiers, using contrastive learning methods, etc. Nevertheless, they still rely on a large amount of target domain data to function well. Instead, we address this problem by establishing a concrete performance bound of a policy given datasets from two domains. Motivated by the theoretical insights, we propose to align transitions in the two datasets using optimal transport and selectively share source domain samples, without training any neural networks. This enables reliable data filtering even given a few target domain data. Additionally, we introduce a dataset regularization term that ensures the learned policy remains within the scope of the target domain dataset, preventing it from being biased towards the source domain data. Consequently, we propose the Optimal Transport Data Filtering (dubbed OTDF) method and examine its effectiveness by conducting extensive experiments across various dynamics shift conditions (e.g., gravity shift), given limited target domain data. It turns out that OTDF exhibits superior performance on many tasks and dataset qualities, often surpassing prior strong baselines by a large margin.

Rectified-flow-based diffusion transformers like FLUX and OpenSora have demonstrated outstanding performance in the field of image and video generation. Despite their robust generative capabilities, these models often struggle with inversion inaccuracies, which could further limit their effectiveness in downstream tasks such as image and video editing. To address this issue, we propose RF-Solver, a novel training-free sampler that effectively enhances inversion precision by mitigating the errors in the ODE-solving process of rectified flow. Specifically, we derive the exact formulation of the rectified flow ODE and apply the high-order Taylor expansion to estimate its nonlinear components, significantly enhancing the precision of ODE solutions at each timestep. Building upon RF-Solver, we further propose RF-Edit, a general feature-sharing-based framework for image and video editing. By incorporating self-attention features from the inversion process into the editing process, RF-Edit effectively preserves the structural information of the source image or video while achieving high-quality editing results. Our approach is compatible with any pre-trained rectified-flow-based models for image and video tasks, requiring no additional training or optimization. Extensive experiments across generation, inversion, and editing tasks in both image and video modalities demonstrate the superiority and versatility of our method. The source code is available at https://github.com/wangjiangshan0725/RF-Solver-Edit.

Recent advances have highlighted the benefits of scaling language models to enhance performance across a wide range of NLP tasks. However, these approaches still face limitations in effectiveness and efficiency when applied to domain-specific tasks, particularly for small edge-side models. We propose the LoRA-Gen framework, which utilizes a large cloud-side model to generate LoRA parameters for edge-side models based on task descriptions. By employing the reparameterization technique, we merge the LoRA parameters into the edge-side model to achieve flexible specialization. Our method facilitates knowledge transfer between models while significantly improving the inference efficiency of the specialized model by reducing the input context length. Without specialized training, LoRA-Gen outperforms conventional LoRA fine-tuning, which achieves competitive accuracy and a 2.1x speedup with TinyLLaMA-1.1B in reasoning tasks. Besides, our method delivers a compress ratio of 10.1x with Gemma-2B on intelligent agent tasks.

In real-world recommendation scenarios, users engage with items through various types of behaviors. Leveraging diversified user behavior information for learning can enhance the recommendation of target behaviors (e.g., buy), as demonstrated by recent multi-behavior methods. The mainstream multi-behavior recommendation framework consists of two steps: fusion and prediction. Recent approaches utilize graph neural networks for multi-behavior fusion and employ multi-task learning paradigms for joint optimization in the prediction step, achieving significant success. However, these methods have limited perspectives on multi-behavior fusion, which leads to inaccurate capture of user behavior patterns in the fusion step. Moreover, when using multi-task learning for prediction, the relationship between the target task and auxiliary tasks is not sufficiently coordinated, resulting in negative information transfer. To address these problems, we propose a novel multi-behavior recommendation framework based on the combinatorial optimization perspective, named COPF. Specifically, we treat multi-behavior fusion as a combinatorial optimization problem, imposing different constraints at various stages of each behavior to restrict the solution space, thus significantly enhancing fusion efficiency (COGCN). In the prediction step, we improve both forward and backward propagation during the generation and aggregation of multiple experts to mitigate negative transfer caused by differences in both feature and label distributions (DFME). Comprehensive experiments on three real-world datasets indicate the superiority of COPF. Further analyses also validate the effectiveness of the COGCN and DFME modules. Our code is available at https://github.com/1918190/COPF.

Existing LLM-based motion models fail to fully leverage large models' planning capabilities for motion-related tasks, exhibiting poor generalization, limited text-motion alignment, and an inability to perform multimodal condition joint driven motion generation. We argue that these issues arise from the modality gap and the highly coupled nature of motion tokens. To address this, we proposed the hybrid motion sentence, which is consistant of fine-grained motion decription and atomic body-part motion token that can bridge the gap between motion and text. To obtain a large corpus of hybrid motion sentences, we introduced a novel motion-to-text generation method that combines atomic motion operators with GPT-4o, resulting in 68.2 million fine-grained textual descriptions across diverse modalities. To reconstruct high-quality motion from hybrid sentences and make better motion-text alignment, we introduce Semantic-Aware Decoupled Motion Tokenization. Furthermore, we propose MotionUPG based on LLaMA, leveraging MotionWords dataset for both pretraining and instruction tuning. Our method achieves strong fine-grained text-motion alignment, impressive zero-shot motion generation, and is the first to support multimodal condition joint driven motion generation tasks.

The success of diffusion models in text-to-image (T2I) generation has made it urgent to remove unwanted concepts, such as copyrighted, offensive, and unsafe ones, from pre-trained models in an accurate, timely, and cost-effective manner. However, limited by the inherent optimization perspective, existing methods have two major problems. Firstly, they overlook maintaining the global visual style during the erasure process, leading to significant style shifts. Secondly, excessive concept erasure causes relevant content to disappear or generates substitutes unrelated to the original object's attributes. Compared to other methods, our proposed ICE has unique advantages, as it can generate diverse visual features and achieve a balance between concept erasure and maintaining the semantic content of the target object. This is mainly achieved through our well-designed non-erasable features protector (NEFP) and augmented invariant constraints (AIC). Specifically, we enhance the protection of feature information by embedding an augmented orthogonal anchor concept matrix. Meanwhile, under controlled constraints, we introduce invariants into the embedding space to retain key semantics. This work specifically emphasizes the importance of focusing on feature expression and semantic protection in the concept erasure task for fully unleashing the performance of T2I models.

Image assessment aims to evaluate the quality and aesthetics of images and has been applied across various scenarios, such as natural and AIGC scenes. Existing methods mostly address these sub-tasks or scenes individually. While some works attempt to develop unified image assessment models, they have struggled to achieve satisfactory performance or cover a broad spectrum of assessment scenarios. In this paper, we present Gamma, a Generic imAge assessMent model using Mixture of Assessment Experts, which can effectively assess images from diverse scenes through mixed-dataset training. Achieving unified training in image assessment presents significant challenges due to annotation biases across different datasets. To address this issue, we first propose a Mixture of Assessment Experts (MoAE) module, which employs shared and adaptive experts to dynamically learn common and specific knowledge for different datasets, respectively. In addition, we introduce a Scene-based Differential Prompt (SDP) strategy, which uses scene-specific prompts to provide prior knowledge and guidance during the learning process, further boosting adaptation for various scenes. Our Gamma model is trained and evaluated on 12 datasets spanning 6 image assessment scenarios. Extensive experiments show that our unified Gamma outperforms other state-of-the-art mixed-training methods by significant margins while covering more scenes. Codes are available at https://github.com/zht8506/Gamma.

Zeyuan Liu, Ziyu Huan, Xiyao Wang, Jiafei Lyu, Jian Tao, Xiu Li, Furong Huang, Huazhe Xu. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Leveraging multimodal large models for image segmentation has become a prominent research direction. However, existing approaches typically rely heavily on manually annotated datasets that include explicit reasoning processes, which are costly and time-consuming to produce. Recent advances suggest that reinforcement learning (RL) can endow large models with reasoning capabilities without requiring such reasoning-annotated data. In this paper, we propose SAM-R1, a novel framework that enables multimodal large models to perform fine-grained reasoning in image understanding tasks. Our approach is the first to incorporate fine-grained segmentation settings during the training of multimodal reasoning models. By integrating task-specific, fine-grained rewards with a tailored optimization objective, we further enhance the model's reasoning and segmentation alignment. We also leverage the Segment Anything Model (SAM) as a strong and flexible reward provider to guide the learning process. With only 3k training samples, SAM-R1 achieves strong performance across multiple benchmarks, demonstrating the effectiveness of reinforcement learning in equipping multimodal models with segmentation-oriented reasoning capabilities.

Real-world datasets collected from sensors or human inputs are prone to noise and errors, posing significant challenges for applying offline reinforcement learning (RL). While existing methods have made progress in addressing corrupted actions and rewards, they remain insufficient for handling corruption in high-dimensional state spaces and for cases where multiple elements in the dataset are corrupted simultaneously. Diffusion models, known for their strong denoising capabilities, offer a promising direction for this problem—but their tendency to overfit noisy samples limits their direct applicability. To overcome this, we propose **A**mbient **D**iffusion-**G**uided Dataset Recovery (**ADG**), a novel approach that pioneers the use of diffusion models to tackle data corruption in offline RL. First, we introduce Ambient Denoising Diffusion Probabilistic Models (DDPM) from approximated distributions, which enable learning on partially corrupted datasets with theoretical guarantees. Second, we use the noise-prediction property of Ambient DDPM to distinguish between clean and corrupted data, and then use the clean subset to train a standard DDPM. Third, we employ the trained standard DDPM to refine the previously identified corrupted data, enhancing data quality for subsequent offline RL training. A notable strength of ADG is its versatility—it can be seamlessly integrated with any offline RL algorithm. Experiments on a range of benchmarks, including MuJoCo, Kitchen, and Adroit, demonstrate that ADG effectively mitigates the impact of corrupted data and improves the robustness of offline RL under various noise settings, achieving state-of-the-art results.

Vision Transformers (ViTs) have become a universal backbone for both image recognition and image generation. Yet their Multi–Head Self–Attention (MHSA) layer still performs a quadratic query–key interaction for \emph{every} token pair, spending the bulk of computation on visually weak or redundant correlations. We introduce \emph{Visual–Contrast Attention} (VCA), a drop-in replacement for MHSA that injects an explicit notion of discrimination while reducing the theoretical complexity from $\mathcal{O}(N^{2}C)$ to $\mathcal{O}(N n C)$ with $n\!\ll\!N$. VCA first distils each head’s dense query field into a handful of spatially pooled \emph{visual–contrast tokens}, then splits them into a learnable \emph{positive} and \emph{negative} stream whose differential interaction highlights what truly separates one region from another. The module adds fewer than $0.3$\,M parameters to a DeiT-Tiny backbone, requires no extra FLOPs, and is wholly architecture-agnostic. Empirically, VCA lifts DeiT-Tiny top-1 accuracy on ImageNet-1K from $72.2\%$ to \textbf{$75.6\%$} (+$3.4$) and improves three strong hierarchical ViTs by up to $3.1$\%, while in class-conditional ImageNet generation it lowers FID-50K by $2.1$ to $5.2$ points across both diffusion (DiT) and flow (SiT) models. Extensive ablations confirm that (i) spatial pooling supplies low-variance global cues, (ii) dual positional embeddings are indispensable for contrastive reasoning, and (iii) combining the two in both stages yields the strongest synergy. VCA therefore offers a simple path towards faster and sharper Vision Transformers. The source code is available at \href{https://github.com/LeapLabTHU/LinearDiff}{https://github.com/LeapLabTHU/LinearDiff}.

Modern interactive applications increasingly demand dynamic 3D content, yet the transformation of static 3D models into animated assets constitutes a significant bottleneck in content creation pipelines. While recent advances in generative AI have revolutionized static 3D model creation, rigging and animation continue to depend heavily on expert intervention. We present \textbf{Puppeteer}, a comprehensive framework that addresses both automatic rigging and animation for diverse 3D objects. Our system first predicts plausible skeletal structures via an auto-regressive transformer that introduces a joint-based tokenization strategy for compact representation and a hierarchical ordering methodology with stochastic perturbation that enhances bidirectional learning capabilities. It then infers skinning weights via an attention-based architecture incorporating topology-aware joint attention that explicitly encodes inter-joint relationships based on skeletal graph distances. Finally, we complement these rigging advances with a differentiable optimization-based animation pipeline that generates stable, high-fidelity animations while being computationally more efficient than existing approaches. Extensive evaluations across multiple benchmarks demonstrate that our method significantly outperforms state-of-the-art techniques in both skeletal prediction accuracy and skinning quality. The system robustly processes diverse 3D content, ranging from professionally designed game assets to AI-generated shapes, producing temporally coherent animations that eliminate the jittering issues common in existing methods.

Recent text-to-image systems face limitations in handling multimodal inputs and complex reasoning tasks. We introduce MindOmni, a unified multimodal large language model that addresses these challenges by incorporating reasoning generation through reinforcement learning. MindOmni leverages a three-phase training strategy: i) design of a unified vision language model with a decoder-only diffusion module, ii) supervised fine-tuning with Chain-of-Thought (CoT) instruction data, and iii) our proposed Reasoning Generation Policy Optimization (RGPO) algorithm, utilizing multimodal feedback to effectively guide policy updates. Experimental results demonstrate that MindOmni outperforms existing models, achieving impressive performance on both understanding and generation benchmarks, meanwhile showcasing advanced fine-grained reasoning generation capabilities, especially with mathematical reasoning instruction. All codes will be made public.

World models have recently attracted growing interest in Multi-Agent Reinforcement Learning (MARL) due to their ability to improve sample efficiency for policy learning. However, accurately modeling environments in MARL is challenging due to the exponentially large joint action space and highly uncertain dynamics inherent in multi-agent systems. To address this, we reduce modeling complexity by shifting from jointly modeling the entire state-action transition dynamics to focusing on the state space alone at each timestep through sequential agent modeling. Specifically, our approach enables the model to progressively resolve uncertainty while capturing the structured dependencies among agents, providing a more accurate representation of how agents influence the state. Interestingly, this sequential revelation of agents' actions in a multi-agent system aligns with the reverse process in diffusion models—a class of powerful generative models known for their expressiveness and training stability compared to autoregressive or latent variable models. Leveraging this insight, we develop a flexible and robust world model for MARL using diffusion models. Our method, \textbf{D}iffusion-\textbf{I}nspired \textbf{M}ulti-\textbf{A}gent world model (DIMA), achieves state-of-the-art performance across multiple multi-agent control benchmarks, significantly outperforming prior world models in terms of final return and sample efficiency, including MAMuJoCo and Bi-DexHands. DIMA establishes a new paradigm for constructing multi-agent world models, advancing the frontier of MARL research. Codes are open-sourced at \url{https://github.com/breez3young/DIMA}.

The popularity of short video applications has brought new opportunities and challenges to video recommendation.In addition to the traditional ranking-based pipeline, industrial solutions usually introduce additional distribution management components to guarantee a diverse and content-rich user experience.However, existing solutions are either non-personalized or fail to generalize well to the ever-changing user preferences.Inspired by the success of uplift modeling in online marketing, we attempt to implement uplift modeling in the video recommendation scenario to mitigate the problems.However, we face two main challenges when migrating the technique: 1) the complex-response causal relation in distribution management problem, and 2) the modeling of long-term and real-time user preferences.To address these challenges, we correspond each treatment to a specific adjustment of the distribution over video types, then propose a Coarse-to-fine Dynamic Uplift Modeling (CDUM) framework for real-time video recommendation scenarios.Specifically, CDUM consists of two modules, a coarse-grained module that utilizes the offline features of users to model their long-term preferences, and a fine-grained * Both authors contributed equally to this research.

Abstract Multi-camera depth estimation has gained significant attention in autonomous driving due to its importance in perceiving complex environments. However, extending monocular self-supervised methods to multi-camera setups introduces unique challenges that existing techniques often fail to address. In this paper, we propose STViT+ , a novel Transformer-based framework for self-supervised multi-camera depth estimation. Our key contributions include: 1) the Spatial-Temporal Transformer (STTrans) , which integrates local spatial connectivity and global context to capture enriched spatial-temporal cross-view correlations, resulting in more accurate 3D geometry reconstruction; 2) the Spatial-Temporal Photometric Consistency Correction (STPCC) strategy that mitigates the impact of varying illumination, ensuring brightness consistency across frames during photometric loss calculation; 3) the Adversarial Geometry Regularization (AGR) module, which employs Generative Adversarial Networks to impose spatial constraints by using unpaired depth maps, enhancing performance under adverse conditions such as rain and nighttime driving. Extensive evaluations on large-scale autonomous driving datasets, including Nuscenes and DDAD, confirm that STViT+ sets a new benchmark for multi-camera depth estimation.

Narrative economics suggests that financial markets are strongly influenced by evolving narratives, creating opportunities for forecasting emerging events and their economic impacts. However, existing large language model (LLM)-based approaches are inadequate in terms of systematic task decomposition and alignment with financial applications. We propose MENTOR, a multi-agent framework for event and narrative trend prediction that integrates teacher-student iterative reasoning with progressive subtasks: detecting and ranking trending events, forecasting future events from current narratives, and predicting industry index performance influenced by these events. Experiments on our self-constructed Chinese key opinion leader (KOL) articles dataset and English financial news dataset show that MENTOR consistently outperforms recent baselines such as the stakeholder-enhanced future event prediction (StkFEP) and summarize–explain–predict (SEP) frameworks in both event prediction and industry ranking tasks. In addition, the backtest results at the portfolio level show that improved event and industry forecasts can bring about a practical improvement in investment performance. These results demonstrate that incorporating structured reasoning and multi-agent feedback enables more reliable event forecasting and strengthens the connection between narrative dynamics and financial market outcomes.

How the recent progress of reasoning large language models (LLMs), especially the new open-source model DeepSeek-R1, can benefit financial services is an underexplored problem. While LLMs have ignited numerous applications within the financial sector, including financial news analysis and general customer interactions, DeepSeek-R1 further unlocks the advanced reasoning ability with multiple reinforcement learning-integrated training steps for more complex financial queries and provides distilled student models for resource-constrained scenarios. In this paper, we first introduce the technological preliminaries of DeepSeek-R1. Subsequently, we benchmark the performance of DeepSeek-R1 and its distilled students on two public financial question–answer (QA) datasets as a starting point for interdisciplinary research on financial artificial intelligence (AI). Then, we discuss the opportunities that DeepSeek-R1 offers to current financial services, its current limitations, and three future research directions. In conclusion, we argue for a proper approach to adopting reasoning LLMs for financial AI.

Incompletely-Supervised Concealed Object Segmentation (ISCOS) involves segmenting objects that seamlessly blend into their surrounding environments, utilizing incompletely annotated data, such as weak and semi-annotations, for model training. This task remains highly challenging due to (1) the limited supervision provided by the incompletely annotated training data, and (2) the difficulty of distinguishing concealed objects from the background, which arises from the intrinsic similarities in concealed scenarios. In this paper, we introduce the first unified method for ISCOS to address these challenges. To tackle the issue of incomplete supervision, we propose a unified mean-teacher framework, SEE, that leverages the vision foundation model, "Segment Anything Model (SAM)", to generate pseudo-labels using coarse masks produced by the teacher model as prompts. To mitigate the effect of low-quality segmentation masks, we introduce a series of strategies for pseudo-label generation, storage, and supervision. These strategies aim to produce informative pseudo-labels, store the best pseudo-labels generated, and select the most reliable components to guide the student model, thereby ensuring robust network training. Additionally, to tackle the issue of intrinsic similarity, we design a hybrid-granularity feature grouping module that groups features at different granularities and aggregates these results. By clustering similar features, this module promotes segmentation coherence, facilitating more complete segmentation for both single-object and multiple-object images. We validate the effectiveness of our approach across multiple ISCOS tasks, and experimental results demonstrate that our method achieves state-of-the-art performance. Furthermore, SEE can serve as a plug-and-play solution, enhancing the performance of existing models.

Deep generative models have gained considerable attention in low-level vision tasks due to their powerful generative capabilities. Among these, diffusion model-based approaches, which employ a forward diffusion process to degrade an image and a reverse denoising process for image generation, have become particularly prominent for producing high-quality, diverse samples with intricate texture details. Despite their widespread success in low-level vision, there remains a lack of a comprehensive, insightful survey that synthesizes and organizes the advances in diffusion model-based techniques. To address this gap, this paper presents the first comprehensive review focused on denoising diffusion models applied to low-level vision tasks, covering both theoretical and practical contributions. We outline three general diffusion modeling frameworks and explore their connections with other popular deep generative models, establishing a solid theoretical foundation for subsequent analysis. We then categorize diffusion models used in low-level vision tasks from multiple perspectives, considering both the underlying framework and the target application. Beyond natural image processing, we also summarize diffusion models applied to other low-level vision domains, including medical imaging, remote sensing, and video processing. Additionally, we provide an overview of widely used benchmarks and evaluation metrics in low-level vision tasks. Our review includes an extensive evaluation of diffusion model-based techniques across six representative tasks, with both quantitative and qualitative analysis. Finally, we highlight the limitations of current diffusion models and propose four promising directions for future research. This comprehensive review aims to foster a deeper understanding of the role of denoising diffusion models in low-level vision.

Recent years have witnessed the rapid development of general human action understanding. However, when applied to real-world applications such as sports analysis, most existing datasets are still unsatisfactory, because of the limitations in rich labels on multiple tasks, language instructions, high-quality 3D data, and diverse environments. In this paper, we present FLAG3D++, a large-scale benchmark for 3D fitness activity comprehension, which contains 180 K sequences of 60 activity categories with language instruction. FLAG3D++ features the following four aspects: 1) fine-grained annotations of the temporal intervals of actions in the untrimmed long sequences and how well these actions are performed, 2) detailed and professional language instruction to describe how to perform a specific activity, 3) accurate and dense 3D human pose captured from advanced MoCap system to handle the complex activity and large movement, 4) versatile video resources from a high-tech MoCap system, rendering software, and cost-effective smartphones in natural environments. In light of the specified features, we present two new practical applications as language-guided repetition action counting (L-RAC) and language-guided action quality assessment (L-AQA), which aim to take the language descriptions as references to count the repetitive times of an action and assess the quality of action respectively. Furthermore, we propose a Hierarchical Language-Guided Graph Convolutional Network (HL-GCN) model to better fuse the language information and skeleton sequences for L-RAC and L-AQA. To be specific, the HL-GCN performs cross-modal alignments by the early fusion of the linguistic feature and the hierarchical node features of the skeleton-based sequences encoded by the multiple intermediate graph convolutional layers. Extensive experiments show the superiority of our HL-GCN on both L-RAC and L-AQA, as well as the great research value of FLAG3D++ for various challenges, such as dynamic human mesh recovery and cross-domain human action recognition. Our dataset, source code, and trained models are made publicly available at FLAG3D++.

Radiographic images are similar to each other, making it challenging for diagnostic captioning to narrate fine-grained visual differences of clinical importance. In this paper, we propose a self-boosting framework integrating two novel strategies to learn tightly correlated image and text features for diagnostic captioning. The first strategy explicitly aligns image and text features through training an auxiliary task of image-text matching (ITM) jointly with the main task of report generation (RG) as two branches of a network model. The ITM branch explicitly learns image-text alignment and provides highly correlated visual and textual features for the RG branch to generate high-quality reports. The high-quality reports generated by RG branch, in turn, are utilized as additional harder negative samples to push the ITM branch to evolve towards better image-text alignment. These two branches help improve each other progressively, so that the whole model is self-boosted without requiring external resources. The second strategy aligns image-sample space and report-sample space to achieve consistent image and text feature embeddings. To achieve this, the sample graph of the embedded ground-truth reports is built and used as the target to train the sample graph of the embedded images so that the fine discrepancy in the ground-truth reports could be captured by the learned visual feature embeddings. Our proposed framework demonstrates its superiority on two medical report generation benchmarks, including the largest dataset MIMIC-CXR.

The detection head constitutes a pivotal component within object detectors, tasked with executing both classification and localization functions. Regrettably, the commonly used parallel head often lacks omni perceptual capabilities, such as deformation perception (DP), global perception (GP), and cross-task perception (CTP). Despite numerous methods attempting to enhance these abilities from a single aspect, achieving a comprehensive and unified solution remains a significant challenge. In response to this challenge, we develop an innovative detection head, termed UniHead, to unify three perceptual abilities simultaneously. More precisely, our approach: 1) introduces DP, enabling the model to adaptively sample object features; 2) proposes a dual-axial aggregation transformer (DAT) to adeptly model long-range dependencies, thereby achieving GP; and 3) devises a cross-task interaction transformer (CIT) that facilitates interaction between the classification and localization branches, thus aligning the two tasks. As a plug-and-play method, the proposed UniHead can be conveniently integrated with existing detectors. Extensive experiments on the COCO dataset demonstrate that our UniHead can bring significant improvements to many detectors. For instance, the UniHead can obtain +2.7 AP gains in RetinaNet, +2.9 AP gains in FreeAnchor, and +2.1 AP gains in GFL. The code is available at https://github.com/zht8506/UniHead.

Although remarkable progress has been made in image style transfer, style is just one of the components of artistic paintings. Directly transferring extracted style features to natural images often results in outputs with obvious synthetic traces. This is because key painting attributes including layout, perspective, shape, and semantics often cannot be conveyed and expressed through style transfer. Large-scale pretrained text-to-image generation models have demonstrated their capability to synthesize a vast amount of high-quality images. However, even with extensive textual descriptions, it is challenging to fully express the unique visual properties and details of paintings. Moreover, generic models often disrupt the overall artistic effect when modifying specific areas, making it more complicated to achieve a unified aesthetic in artworks. Our main novel idea is to integrate multimodal semantic information as a synthesis guide into artworks, rather than transferring style to the real world. We also aim to reduce the disruption to the harmony of artworks while simplifying the guidance conditions. Specifically, we propose an innovative multi-task unified framework called CreativeSynth, based on the diffusion model with the ability to coordinate multimodal inputs. CreativeSynth combines multimodal features with customized attention mechanisms to seamlessly integrate real-world semantic content into the art domain through Cross-Art-Attention for aesthetic maintenance and semantic fusion. We demonstrate the results of our method across a wide range of different art categories, proving that CreativeSynth bridges the gap between generative models and artistic expression.

3D-aware GANs successfully solve the problem of 3D-consistency generation and furthermore provide a 3D shape of the generated object. However, the application of the volume renderer disturbs the disentanglement of the latent space, which makes it difficult to manipulate 3D-aware GANs and lowers the image quality of style-based generators. In this work, we devise a dual-mapping framework to make the generated images of pretrained 2D StyleGAN consistent in 3D space. We utilize a tri-plane representation to estimate the 3D shape of the generated object and two mapping networks to bridge the latent space of StyleGAN and the 3D tri-plane space. Our method does not alter the parameters of the pretrained generator, which means the interpretability of latent space is preserved for various image manipulations. Experiments show that our method lifts the 3D awareness of pretrained 2D StyleGAN to 3D-aware GANs and outperforms the 3D-aware GANs in controllability and image quality.

Multi-camera depth estimation has recently garnered significant attention due to its substantial practical implications in the realm of autonomous driving. In this paper, we delve into the task of self-supervised multi-camera depth estimation and propose an innovative framework, STViT, featuring several noteworthy enhancements: 1) we propose a Spatial-Temporal Transformer to comprehensively exploit both local connectivity and the global context of image features, meanwhile learning enriched spatial-temporal cross-view correlations to recover 3D geometry. 2) to alleviate the severe effect of adverse conditions, e.g., rainy weather and nighttime driving, we introduce a GAN-based Adversarial Geometry Regularization Module (AGR) to further constrain the depth estimation with unpaired normal-condition depth maps and prevent the model from being incorrectly trained. Experiments on challenging autonomous driving datasets Nuscenes and DDAD show that our method achieves state-of-the-art performance.

Generating text-editable and pose-controllable character videos have an imperious demand in creating various digital human. Nevertheless, this task has been restricted by the absence of a comprehensive dataset featuring paired video-pose captions and the generative prior models for videos. In this work, we design a novel two-stage training scheme that can utilize easily obtained datasets (i.e., image pose pair and pose-free video) and the pre-trained text-to-image (T2I) model to obtain the pose-controllable character videos. Specifically, in the first stage, only the keypoint image pairs are used only for a controllable text-to-image generation. We learn a zero-initialized convolutional encoder to encode the pose information. In the second stage, we finetune the motion of the above network via a pose-free video dataset by adding the learnable temporal self-attention and reformed cross-frame self-attention blocks. Powered by our new designs, our method successfully generates continuously pose-controllable character videos while keeps the editing and concept composition ability of the pre-trained T2I model. The code and models are available on https://follow-your-pose.github.io/.

This study aims to improve the generation of 3D gestures by utilizing multimodal information from human speech. Previous studies have focused on incorporating additional modalities to enhance the quality of generated gestures. However, these methods perform poorly when certain modalities are missing during inference. To address this problem, we suggest using speech-derived multimodal priors to improve gesture generation. We introduce a novel method that separates priors from speech and employs multimodal priors as constraints for generating gestures. Our approach utilizes a chain-like modeling method to generate facial blendshapes, body movements, and hand gestures sequentially. Specifically, we incorporate rhythm cues derived from facial deformation and stylization prior based on speech emotions, into the process of generating gestures. By incorporating multimodal priors, our method improves the quality of generated gestures and eliminate the need for expensive setup preparation during inference. Extensive experiments and user studies confirm that our proposed approach achieves state-of-the-art performance.

Language conditioned 3D object grounding aims to find the object within the 3D scene mentioned by natural language descriptions, which mainly depends on the matching between visual and natural language. Considerable improvement in grounding performance is achieved by improving the multimodal fusion mechanism or bridging the gap between detection and matching. However, several mismatches are ignored, i.e., mismatch in local visual representation and global sentence representation, and mismatch in visual space and corresponding label word space. In this paper, we propose crossmodal match for 3D grounding from mitigating these mismatches perspective. Specifically, to match local visual features with the global description sentence, we propose BEV (Bird’s-eye-view) based global information embedding module. It projects multiple object proposal features into the BEV and the relations of different objects are accessed by the visual transformer which can model both positions and features with long-range dependencies. To circumvent the mismatch in feature spaces of different modalities, we propose crossmodal consistency learning. It performs cross-modal consistency constraints to convert the visual feature space into the label word feature space resulting in easier matching. Besides, we introduce label distillation loss and global distillation loss to drive these matches learning in a distillation way. We evaluate our method in mainstream evaluation settings on three datasets, and the results demonstrate the effectiveness of the proposed method.

We propose Lodge, a network capable of generating extremely long dance sequences conditioned on given music. We design Lodge as a two-stage coarse to fine diffusion architecture, and propose the characteristic dance primitives that possess significant expressiveness as intermediate representations between two diffusion models. The first stage is global diffusion, which focuses on comprehending the coarse-level music-dance correlation and production characteristic dance primitives. In contrast, the second-stage is the local diffusion, which parallelly generates detailed motion sequences under the guidance of the dance primitives and choreographic rules. In addition, we propose a Foot Refine Block to optimize the contact between the feet and the ground, enhancing the physical realism of the motion. Our approach can parallelly generate dance sequences of extremely long length, striking a balance between global choreographic patterns and local motion quality and expressiveness. Extensive experiments validate the efficacy of our method. Code, models, and demonstrative video results are available at: https://li-ronghui.github.io/lodge

Video Moment Retrieval (MR) and Highlight Detection (HD) have attracted significant attention due to the growing demand for video analysis. Recent approaches treat MR and HD as similar video grounding problems and address them together with transformer-based architecture. However, we observe that the emphasis of MR and HD differs, with one necessitating the perception of local relationships and the other prioritizing the understanding of global contexts. Consequently, the lack of task-specific design will inevitably lead to limitations in associating the intrinsic specialty of two tasks. To tackle the issue, we propose a Unified Video COMprehension framework (UVCOM) to bridge the gap and jointly solve MR and HD effectively. By performing progressive integration on intra and inter-modality across multi-granularity, UVCOM achieves the comprehensive understanding in processing a video. Moreover, we present multi-aspect contrastive learning to consolidate the local relation modeling and global knowledge accumulation via well aligned multi-modal space. Extensive experiments on QVHighlights, Charades-STA, TACoS, YouTube Highlights and TVSum datasets demonstrate the effectiveness and rationality of UVCOM which outperforms the state-of-the-art methods by a remarkable margin. Code is available at https://github.com/EasonXiao-888/UVCOM.

Using reinforcement learning with human feedback (RLHF) has shown significant promise in fine-tuning diffusion models. Previous methods start by training a reward model that aligns with human preferences, then leverage RL techniques to fine-tune the underlying models. However, crafting an efficient reward model demands extensive datasets, optimal architecture, and manual hyperparameter tuning, making the process both time and cost-intensive. The direct preference optimization (DPO) method, effective in fine-tuning large language models, eliminates the necessity for a reward model. However, the extensive GPU memory requirement of the diffusion model's denoising process hinders the direct application of the DPO method. To address this issue, we introduce the Direct Preference for De-noising Diffusion Policy Optimization (D3PO) method to directly fine-tune diffusion models. The theoretical analysis demonstrates that although D3PO omits training a reward model, it effectively functions as the optimal re-ward model trained using human feedback data to guide the learning process. This approach requires no training of a reward model, proving to be more direct, cost-effective, and minimizing computational overhead. In experiments, our method uses the relative scale of objectives as a proxy for human preference, delivering comparable results to methods using ground-truth rewards. Moreover, D3PO demonstrates the ability to reduce image distortion rates and generate safer images, overcoming challenges lacking robust reward models. Our code is publicly available at https://github.com/yk7333/D3PO.

The primacy bias in model-free reinforcement learning (MFRL), which refers to the agent’s tendency to overfit early data and lose the ability to learn from new data, can significantly decrease the performance of MFRL algorithms. Previous studies have shown that employing simple techniques, such as resetting the agent’s parameters, can substantially alleviate the primacy bias in MFRL. However, the primacy bias in model-based reinforcement learning (MBRL) remains unexplored. In this work, we focus on investigating the primacy bias in MBRL. We begin by observing that resetting the agent’s parameters harms its performance in the context of MBRL. We further find that the primacy bias in MBRL is more closely related to the primacy bias of the world model instead of the primacy bias of the agent. Based on this finding, we propose world model resetting, a simple yet effective technique to alleviate the primacy bias in MBRL. We apply our method to two different MBRL algorithms, MBPO and DreamerV2. We validate the effectiveness of our method on multiple continuous control tasks on MuJoCo and DeepMind Control Suite, as well as discrete control tasks on Atari 100k benchmark. The experimental results show that world model resetting can significantly alleviate the primacy bias in the model-based setting and improve the algorithm’s performance. We also give a guide on how to perform world model resetting effectively.

Generating 3D human models directly from text helps reduce the cost and time of character modeling. However, achieving multi-attribute controllable and realistic 3D human avatar generation is still challenging due to feature coupling and the scarcity of realistic 3D human avatar datasets. To address these issues, we propose Text2Avatar, which can generate realistic-style 3D avatars based on the coupled text prompts. Text2Avatar leverages a discrete codebook as an intermediate feature to establish a connection between text and avatars, enabling the disentanglement of features. Furthermore, to alleviate the scarcity of realistic style 3D human avatar data, we utilize a pre-trained unconditional 3D human avatar generation model to obtain a large amount of 3D avatar pseudo data, which allows Text2Avatar to achieve realistic style generation. Experimental results demonstrate that our method can generate realistic 3D avatars from coupled textual data, which is challenging for other existing methods in this field.

Existing music-driven 3D dance generation methods mainly concentrate on high-quality dance generation, but lack sufficient control during the generation process. To address these issues, we propose a unified framework capable of generating high-quality dance movements and supporting multi-modal control, including genre control, semantic control, and spatial control. First, we decouple the dance generation network from the dance control network, thereby avoiding the degradation in dance quality when adding additional control information. Second, we design specific control strategies for different control information and integrate them into a unified framework. Experimental results show that the proposed dance generation framework outperforms state-of-the-art methods in terms of motion quality and controllability.

Camouflaged object detection (COD) is the challenging task of identifying camouflaged objects visually blended into surroundings. Albeit achieving remarkable success, existing COD detectors still struggle to obtain precise results in some challenging cases. To handle this problem, we draw inspiration from the prey-vs-predator game that leads preys to develop better camouflage and predators to acquire more acute vision systems and develop algorithms from both the prey side and the predator side. On the prey side, we propose an adversarial training framework, Camouflageator, which introduces an auxiliary generator to generate more camouflaged objects that are harder for a COD method to detect. Camouflageator trains the generator and detector in an adversarial way such that the enhanced auxiliary generator helps produce a stronger detector. On the predator side, we introduce a novel COD method, called Internal Coherence and Edge Guidance (ICEG), which introduces a camouflaged feature coherence module to excavate the internal coherence of camouflaged objects, striving to obtain more complete segmentation results. Additionally, ICEG proposes a novel edge-guided separated calibration module to remove false predictions to avoid obtaining ambiguous boundaries. Extensive experiments show that ICEG outperforms existing COD detectors and Camouflageator is flexible to improve various COD detectors, including ICEG, which brings state-of-the-art COD performance.

Offline reinforcement learning (RL) has attracted much attention due to its ability in learning from static offline datasets and eliminating the need of interacting with the environment. Nevertheless, the success of offline RL relies heavily on the offline transitions annotated with reward labels. In practice, we often need to hand-craft the reward function, which is sometimes difficult, labor-intensive, or inefficient. To tackle this challenge, we set our focus on the offline imitation learning (IL) setting, and aim at getting a reward function based on the expert data and unlabeled data. To that end, we propose a simple yet effective search-based offline IL method, tagged SEABO. SEABO allocates a larger reward to the transition that is close to its closest neighbor in the expert demonstration, and a smaller reward otherwise, all in an unsupervised learning manner. Experimental results on a variety of D4RL datasets indicate that SEABO can achieve competitive performance to offline RL algorithms with ground-truth rewards, given only a single expert trajectory, and can outperform prior reward learning and offline IL methods across many tasks. Moreover, we demonstrate that SEABO also works well if the expert demonstrations contain only observations. Our code is publicly available at https://github.com/dmksjfl/SEABO.

Storing intermediate frame segmentations as memory for long-range context modeling, spatial-temporal memory-based methods have recently showcased impressive results in semi-supervised video object segmentation (SVOS). However, these methods face two key limitations: 1) relying on non-local pixel-level matching to read memory, resulting in noisy retrieved features for segmentation; 2) segmenting each object independently without interaction. These shortcomings make the memory-based methods struggle in similar object and multi-object segmentation. To address these issues, we propose a query modulation method, termed QMVOS. This method summarizes object features into dynamic queries and then treats them as dynamic filters for mask prediction, thereby providing high-level descriptions and object-level perception for the model. Efficient and effective multi-object interactions are realized through inter-query attention. Extensive experiments demonstrate that our method can bring significant improvements to the memory-based SVOS method and achieve competitive performance on standard SVOS benchmarks. The code is available at https://github.com/zht8506/QMVOS.

In preference-based Reinforcement Learning (RL), obtaining a large number of preference labels are both time-consuming and costly. Furthermore, the queried human preferences cannot be utilized for the new tasks. In this paper, we propose Zero-shot Cross-task Preference Alignment and Robust Reward Learning (PEARL), which learns policies from cross-task preference transfer without any human labels of the target task. Our contributions include two novel components that facilitate the transfer and learning process. The first is Cross-task Preference Alignment (CPA), which transfers the preferences between tasks via optimal transport. The key idea of CPA is to use Gromov-Wasserstein distance to align the trajectories between tasks, and the solved optimal transport matrix serves as the correspondence between trajectories. The target task preferences are computed as the weighted sum of source task preference labels with the correspondence as weights. Moreover, to ensure robust learning from these transferred labels, we introduce Robust Reward Learning (RRL), which considers both reward mean and uncertainty by modeling rewards as Gaussian distributions. Empirical results on robotic manipulation tasks from Meta-World and Robomimic demonstrate that our method is capable of transferring preference labels across tasks accurately and then learns well-behaved policies. Notably, our approach significantly exceeds existing methods when there are few human preferences. The code and videos of our method are available at: https://sites.google.com/view/pearl-preference.

It is vital to learn effective policies that can be transferred to different domains with dynamics discrepancies in reinforcement learning (RL). In this paper, we consider dynamics adaptation settings where there exists dynamics mismatch between the source domain and the target domain, and one can get access to sufficient source domain data, while can only have limited interactions with the target domain. Existing methods address this problem by learning domain classifiers, performing data filtering from a value discrepancy perspective, etc. Instead, we tackle this challenge from a decoupled representation learning perspective. We perform representation learning only in the target domain and measure the representation deviations on the transitions from the source domain, which we show can be a signal of dynamics mismatch. We also show that representation deviation upper bounds performance difference of a given policy in the source domain and target domain, which motivates us to adopt representation deviation as a reward penalty. The produced representations are not involved in either policy or value function, but only serve as a reward penalizer. We conduct extensive experiments on environments with kinematic and morphology mismatch, and the results show that our method exhibits strong performance on many tasks. Our code is publicly available at https://github.com/dmksjfl/PAR.

Exploration remains a critical issue in deep reinforcement learning for an agent to attain high returns in unknown environments. Although the prevailing exploration Random Network Distillation (RND) algorithm has been demonstrated to be effective in numerous environments, it often needs more discriminative power in bonus allocation. This paper highlights the ``bonus inconsistency'' issue within RND, pinpointing its primary limitation. To address this issue, we introduce the Distributional RND (DRND), a derivative of the RND. DRND enhances the exploration process by distilling a distribution of random networks and implicitly incorporating pseudo counts to improve the precision of bonus allocation. This refinement encourages agents to engage in more extensive exploration. Our method effectively mitigates the inconsistency issue without introducing significant computational overhead. Both theoretical analysis and experimental results demonstrate the superiority of our approach over the original RND algorithm. Our method excels in challenging online exploration scenarios and effectively serves as an anti-exploration mechanism in D4RL offline tasks. Our code is publicly available at https://github.com/yk7333/DRND.

Acquiring a multi-task imitation policy in 3D manipulation poses challenges in terms of scene understanding and action prediction. Current methods employ both 3D representation and multi-view 2D representation to predict the poses of the robot’s end-effector. However, they still require a considerable amount of high-quality robot trajectories, and suffer from limited generalization in unseen tasks and inefficient execution in long-horizon reasoning. In this paper, we propose **SAM-E**, a novel architecture for robot manipulation by leveraging a vision-foundation model for generalizable scene understanding and sequence imitation for long-term action reasoning. Specifically, we adopt Segment Anything (SAM) pre-trained on a huge number of images and promptable masks as the foundation model for extracting task-relevant features, and employ parameter-efficient fine-tuning on robot data for a better understanding of embodied scenarios. To address long-horizon reasoning, we develop a novel multi-channel heatmap that enables the prediction of the action sequence in a single pass, notably enhancing execution efficiency. Experimental results from various instruction-following tasks demonstrate that SAM-E achieves superior performance with higher execution efficiency compared to the baselines, and also significantly improves generalization in few-shot adaptation to new tasks.

The robustness of legged locomotion is crucial for quadrupedal robots in challenging terrains. Recently, Reinforcement Learning (RL) has shown promising results in legged locomotion and various methods try to integrate privileged distillation, scene modeling, and external sensors to improve the generalization and robustness of locomotion policies. However, these methods are hard to handle uncertain scenarios such as abrupt terrain changes or unexpected external forces. In this paper, we consider a novel risk-sensitive perspective to enhance the robustness of legged locomotion. Specifically, we employ a distributional value function learned by quantile regression to model the aleatoric uncertainty of environments, and perform risk-averse policy learning by optimizing the worst-case scenarios via a risk distortion measure. Extensive experiments in both simulation environments and a real Aliengo robot demonstrate that our method is efficient in handling various external disturbances, and the resulting policy exhibits improved robustness in harsh and uncertain situations in legged locomotion.

Different types of staining highlight different structures in organs, thereby assisting in diagnosis. However, due to the impossibility of repeated staining, we cannot obtain different types of stained slides of the same tissue area. Translating the slide that is easy to obtain (e.g., H&E) to slides of staining types difficult to obtain (e.g., MT, PAS) is a promising way to solve this problem. However, some regions are closely connected to other regions, and to maintain this connection, they often have complex structures and are difficult to translate, which may lead to wrong translations. In this paper, we propose the Attention-Based Varifocal Generative Adversarial Network (AV-GAN), which solves multiple problems in pathologic image translation tasks, such as uneven translation difficulty in different regions, mutual interference of multiple resolution information, and nuclear deformation. Specifically, we develop an Attention-Based Key Region Selection Module, which can attend to regions with higher translation difficulty. We then develop a Varifocal Module to translate these regions at multiple resolutions. Experimental results show that our proposed AV-GAN outperforms existing image translation methods with two virtual kidney tissue staining tasks and improves FID values by 15.9 and 4.16 respectively in the H&E-MT and H&E-PAS tasks.

Researchers have integrated exploration techniques into multi-agent reinforcement learning (MARL) algorithms, drawing on their remarkable success in deep reinforcement learning. Nonetheless, exploration in MARL presents a more substantial challenge, as agents need to coordinate their efforts to achieve comprehensive state coverage. Reaching a unanimous agreement on which kinds of states warrant exploring can be a struggle for agents in this context. We introduce Multi-agent Exploration based on Sub-state Entropy (MESE ) to address this limitation. This novel approach incentivizes agents to explore states cooperatively by directing them to achieve consensus via an extra team reward. Calculating the additional reward is based on the novelty of the current sub-state that merits cooperative exploration. MESE employs a conditioned entropy approach to select the sub-state, which uses particle-based entropy estimation to calculate the entropy and uses Random Network Distillation(RND) to calculate the team’s intrinsic reward. MESE is a plug-and-play module that can be seamlessly integrated into most existing MARL algorithms, which makes it a highly effective tool for reinforcement learning. Our experiments demonstrate that MESE can substantially improve the performance of MAPPO and QMIX on various tasks in the StarCraft multi-agent challenge (SMAC).

Sparse rewards in reinforcement learning have long been a central research challenge, often tackled through various exploration methods. However, in multi-agent scenarios with larger state spaces and action spaces, sparser rewards, and more complex learning strategies, conventional exploration methods cannot achieve satisfactory results. In this paper, we introduce a novel exploration approach for multi-agent reinforcement learning that combines the strengths of model-based techniques and curiosity-driven methods. We use curiosity-driven exploration in the early stages to facilitate comprehensive exploration of the entire state space. As training progresses, the forward model, trained during the initial exploration phase, is employed to selectively explore crucial actions, allowing agents to discover effective strategies. Due to the decrease in the loss of the curiosity-driven method with an increase in the number of state visits and the increasing accuracy of the forward model with more state visits, we utilize the curiosity-driven loss as a measure of the uncertainty in the forward model. Subsequently, based on the magnitude of this uncertainty, we determine which intrinsic reward to employ. Through experiments in a sparse-reward SMAC environment, we demonstrate the effectiveness of our algorithm. Visualizations of the results further validate the efficacy of our approach in enhancing exploration.

Due to the unique characteristics of underwater environments, accurate 3D reconstruction of underwater objects poses a challenging problem in tasks such as underwater exploration and mapping. Traditional methods that rely on multiple sensor data for 3D reconstruction are time-consuming and face challenges in data acquisition in underwater scenarios. We propose UW-SDF, a framework for reconstructing target objects from multi-view underwater images based on neural SDF. We introduce hybrid geometric priors to optimize the reconstruction process, markedly enhancing the quality and efficiency of neural SDF reconstruction. Additionally, to address the challenge of segmentation consistency in multi-view images, we propose a novel few-shot multi-view target segmentation strategy using the general-purpose segmentation model (SAM), enabling rapid automatic segmentation of unseen objects. Through extensive qualitative and quantitative experiments on diverse datasets, we demonstrate that our proposed method outperforms the traditional underwater 3D reconstruction method and other neural rendering approaches in the field of underwater 3D reconstruction.

Reconstructing 3D objects from a single image guided by pretrained diffusion models has demonstrated promising outcomes. However, due to utilizing the case-agnostic rigid strategy, their generalization ability to arbitrary cases and the 3D consistency of reconstruction are still poor. In this work, we propose Consistent123, a case-aware two-stage method for highly consistent 3D asset reconstruction from one image with both 2D and 3D diffusion priors. In the first stage, Consistent123 utilizes only 3D structural priors for sufficient geometry exploitation, with a CLIP-based case-aware adaptive detection mechanism embedded within this process. In the second stage, 2D texture priors are introduced and progressively take on a dominant guiding role, delicately sculpting the details of the 3D model. Consistent123 aligns more closely with the evolving trends in guidance requirements, adaptively providing adequate 3D geometric initialization and suitable 2D texture refinement for different objects. Consistent123 can obtain highly 3D-consistent reconstruction and exhibits strong generalization ability across various objects. Qualitative and quantitative experiments show that our method significantly outperforms state-of-the-art image-to-3D methods.

Real-world Image Dehazing (RID) aims to alleviate haze-induced degradation in real-world settings. This task remains challenging due to the complexities in accurately modeling real haze distributions and the scarcity of paired real-world data. To address these challenges, we first introduce a cooperative unfolding network that jointly models atmospheric scattering and image scenes, effectively integrating physical knowledge into deep networks to restore haze-contaminated details. Additionally, we propose the first RID-oriented iterative mean-teacher framework, termed the Coherence-based Label Generator, to generate high-quality pseudo labels for network training. Specifically, we provide an optimal label pool to store the best pseudo-labels during network training, leveraging both global and local coherence to select high-quality candidates and assign weights to prioritize haze-free regions. We verify the effectiveness of our method, with experiments demonstrating that it achieves state-of-the-art performance on RID tasks. Code will be available at https://github.com/cnyvfang/CORUN-Colabator.

Widely adopted in modern Vision Transformer designs, Softmax attention can effectively capture long-range visual information; however, it incurs excessive computational cost when dealing with high-resolution inputs. In contrast, linear attention naturally enjoys linear complexity and has great potential to scale up to higher-resolution images. Nonetheless, the unsatisfactory performance of linear attention greatly limits its practical application in various scenarios. In this paper, we take a step forward to close the gap between the linear and Softmax attention with novel theoretical analyses, which demystify the core factors behind the performance deviations. Specifically, we present two key perspectives to understand and alleviate the limitations of linear attention: the injective property and the local modeling ability. Firstly, we prove that linear attention is not injective, which is prone to assign identical attention weights to different query vectors, thus adding to severe semantic confusion since different queries correspond to the same outputs. Secondly, we confirm that effective local modeling is essential for the success of Softmax attention, in which linear attention falls short. The aforementioned two fundamental differences significantly contribute to the disparities between these two attention paradigms, which is demonstrated by our substantial empirical validation in the paper. In addition, more experiment results indicate that linear attention, as long as endowed with these two properties, can outperform Softmax attention across various tasks while maintaining lower computation complexity. Code is available at https://github.com/LeapLabTHU/InLine.

We consider off-dynamics reinforcement learning (RL) where one needs to transfer policies across different domains with dynamics mismatch. Despite the focus on developing dynamics-aware algorithms, this field is hindered due to the lack of a standard benchmark. To bridge this gap, we introduce ODRL, the first benchmark tailored for evaluating off-dynamics RL methods. ODRL contains four experimental settings where the source and target domains can be either online or offline, and provides diverse tasks and a broad spectrum of dynamics shifts, making it a reliable platform to comprehensively evaluate the agent's adaptation ability to the target domain. Furthermore, ODRL includes recent off-dynamics RL algorithms in a unified framework and introduces some extra baselines for different settings, all implemented in a single-file manner. To unpack the true adaptation capability of existing methods, we conduct extensive benchmarking experiments, which show that no method has universal advantages across varied dynamics shifts. We hope this benchmark can serve as a cornerstone for future research endeavors. Our code is publicly available at https://github.com/OffDynamicsRL/off-dynamics-rl.

Video editing is an emerging task, in which most current methods adopt the pre-trained text-to-image (T2I) diffusion model to edit the source video in a zero-shot manner. Despite extensive efforts, maintaining the temporal consistency of edited videos remains challenging due to the lack of temporal constraints in the regular T2I diffusion model. To address this issue, we propose COrrespondence-guided Video Editing (COVE), leveraging the inherent diffusion feature correspondence to achieve high-quality and consistent video editing. Specifically, we propose an efficient sliding-window-based strategy to calculate the similarity among tokens in the diffusion features of source videos, identifying the tokens with high correspondence across frames. During the inversion and denoising process, we sample the tokens in noisy latent based on the correspondence and then perform self-attention within them. To save the usage of GPU memory and accelerate the editing process, we further introduce the temporal-dimensional token merging strategy, which can effectively reduce the redundancy. COVE can be seamlessly integrated into the pre-trained T2I diffusion model without the need for extra training or optimization. Extensive experiment results demonstrate that COVE achieves the start-of-the-art performance in various video editing scenarios, outperforming existing methods both quantitatively and qualitatively. The source code will be released.

The state space models, employing recursively propagated features, demonstrate strong representation capabilities comparable to Transformer models and superior efficiency. However, constrained by the inherent geometric constraints of sequences, it still falls short in modeling long-range dependencies. To address this issue, we propose the MambaTree network, which first dynamically generates a tree topology based on spatial relationships and input features. Then, feature propagation is performed based on this graph, thereby breaking the original sequence constraints to achieve stronger representation capabilities. Additionally, we introduce a linear complexity dynamic programming algorithm to enhance long-range interactions without increasing computational cost. MambaTree is a versatile multimodal framework that can be applied to both visual and textual tasks. Extensive experiments demonstrate that our method significantly outperforms existing structured state space models on image classification, object detection and segmentation. Besides, by fine-tuning large language models, our approach achieves consistent improvements in multiple textual tasks at minor training cost.

Gesture synthesis is a vital realm of human-computer interaction, with wide-ranging applications across various fields like film, robotics, and virtual reality. Recent advancements have utilized the diffusion model to improve gesture synthesis. However, the high computational complexity of these techniques limits the application in reality. In this study, we explore the potential of state space models (SSMs). Direct application of SSMs in gesture synthesis encounters difficulties, which stem primarily from the diverse movement dynamics of various body parts. The generated gestures may also exhibit unnatural jittering issues. To address these, we implement a two-stage modeling strategy with discrete motion priors to enhance the quality of gestures. Built upon the selective scan mechanism, we introduce MambaTalk, which integrates hybrid fusion modules, local and global scans to refine latent space representations. Subjective and objective experiments demonstrate that our method surpasses the performance of state-of-the-art models. Our project is publicly available at~\url{https://kkakkkka.github.io/MambaTalk/}.

Embedding-based retrieval serves as a dominant approach to candidate item matching for industrial recommender systems. With the success of generative AI, generative retrieval has recently emerged as a new retrieval paradigm for recommendation, which casts item retrieval as a generation problem. Its model consists of two stages: semantic tokenization and autoregressive generation. The first stage involves item tokenization that constructs discrete semantic tokens to index items, while the second stage autoregressively generates semantic tokens of candidate items. Therefore, semantic tokenization serves as a crucial preliminary step for training generative recommendation models. Existing research usually employs a vector quantizier with reconstruction loss (e.g., RQ-VAE) to obtain semantic tokens of items, but this method fails to capture the essential neighborhood relationships that are vital for effective item modeling in recommender systems. In this paper, we propose a contrastive quantization-based semantic tokenization approach, named CoST, which harnesses both item relationships and semantic information to learn semantic tokens. Our experimental results highlight the significant impact of semantic tokenization on generative recommendation performance, with CoST achieving up to a 43% improvement in Recall@5 and 44% improvement in NDCG@5 on the MIND dataset over previous baselines.

Click-through rate (CTR) prediction plays a pivotal role in real-world applications, particularly in recommendation systems and online advertising. A significant research branch in this domain focuses on user behavior modeling. Current research predominantly centers on modeling co-occurrence relationships between the target item and items previously interacted with by users. However, this focus neglects the intricate modeling of user behavior patterns. In reality, the abundance of user interaction records encompasses diverse behavior patterns, indicative of a spectrum of habitual paradigms. These patterns harbor substantial potential to significantly enhance CTR prediction performance. To harness the informational potential within behavior patterns, we extend Target Attention (TA) to Target Pattern Attention (TPA) to model pattern-level dependencies. Furthermore, three critical challenges demand attention: the inclusion of unrelated items within patterns, data sparsity of patterns, and computational complexity arising from numerous patterns. To address these challenges, we introduce the Deep Pattern Network (DPN), designed to comprehensively leverage information from behavior patterns. DPN efficiently retrieves target-related behavior patterns using a target-aware attention mechanism. Additionally, it contributes to refining patterns through a pre-training paradigm based on self-supervised learning while promoting dependency learning within sparse patterns. Our comprehensive experiments, conducted across three public datasets, substantiate the superior performance and broad compatibility of DPN.

In the field of dual-target Cross-Domain Recommendation (DTCDR), improving the performance in both the information sparse domain and rich domain has been a mainstream research trend. However, prior embedding-based methods are insufficient to adequately describe the dynamics of user actions and items across domains. Moreover, previous efforts frequently lacked a comprehensive investigation of the entire domain distributions. This paper proposes a novel framework entitled Wasserstein Cross-Domain Recommendation (WCDR) that captures uncertainty in Wasserstein space to address above challenges. In this framework, we abstract user/item actions as Elliptical Gaussian distributions and divide them into local-intrinsic and global-domain parts. To further model the domain diversity, we adopt shared-specific pattern for global-domain distributions and present Masked Domain-aware Sub-distribution Aggregation (MDSA) module to produce informative and diversified global-domain distributions, which incorporates attention-based aggregation method and masking strategy that alleviates negative transfer issues. Extensive experiments on two public datasets and one business dataset are conducted. Experimental results demonstrate the superiority of WCDR over state-of-the-art methods.

For better performance and less resource consumption, multi-scenario recommendation (MSR) is proposed to train a unified model to serve all scenarios by leveraging data from multiple scenarios. Current works in MSR focus on designing effective networks for better information transfer among different scenarios. However, they omit two important issues when applying MSR models in industrial situations. The first is the efficiency problem brought by mixed data, which delays the update of models and further leads to performance degradation. The second is that MSR models are insensitive to the changes of distribution over time, resulting in suboptimal effectiveness in the incoming data. In this paper, we propose an incremental learning approach for MSR (IncMSR), which can not only improve the training efficiency but also perceive changes in distribution over time. Specifically, we first quantify the pair-wise distance between representations from scenario, time and time-scenario dimensions respectively. Then, we decompose the MSR model into scenario-shared and scenario-specific parts and apply fine-grained constraints on the distances quantified with respect to the two different parts. Finally, all constraints are fused in an elegant way using a metric learning framework as a supplementary penalty term to the original MSR loss function. Offline experiments on two real-world datasets are conducted to demonstrate the superiority and compatibility of our proposed approach.

Recently multimodal medical pretraining models play a significant role in automatic medical image and text analysis that has wide social and economical impact in healthcare. Despite being able to be quickly transferred to downstream tasks, the models are greatly limited due to the fact that these models can only be pretrained with professional medical image-text datasets, which usually contain a very small number of samples. In this work We propose MITER (Medical Image-Text Joint adaptive Pretraining), a joint adaptive pretraining framework via multi-level contrastive learning to overcome this limitation by pretraining image and text models for medical domain and utilizing existing models pretrained on generic data, which contain enormous number of samples. MITER features two types of objectives to solve the problem. The first type is uni-modal objectives that pretrain the models with medical images and text separately on uni-modal tasks. The other type is a cross-modal objective that pretrains jointly, allowing the models to influence each other on cross-modal tasks. We also introduce a strategy to dynamically select hard negative samples during the training process for better performance. Experimental results over four medical tasks, image-report retrieval, multi-label image classification, visual question answering, and report generation, show that our MITER framework solves the limitation problem by greatly outperforming existing benchmark models on all the tasks. The source code of our framework is available online.2

Recently, there are many efforts attempting to learn useful policies for continuous control in visual reinforcement learning (RL). In this scenario, it is important to learn a generalizable policy, as the testing environment may differ from the training environment, e.g., there exist distractors during deployment. Many practical algorithms are proposed to handle this problem. However, to the best of our knowledge, none of them provide a theoretical understanding of what affects the generalization gap and why their proposed methods work. In this paper, we bridge this issue by theoretically answering the key factors that contribute to the generalization gap when the testing environment has distractors. Our theories indicate that minimizing the representation distance between training and testing environments, which aligns with human intuition, is the most critical for the benefit of reducing the generalization gap. Our theoretical results are supported by the empirical evidence in the DMControl Generalization Benchmark (DMC-GB).

Unpaired medical image enhancement (UMIE) aims to transform a low-quality (LQ) medical image into a high-quality (HQ) one without relying on paired images for training. While most existing approaches are based on Pix2Pix/CycleGAN and are effective to some extent, they fail to explicitly use HQ information to guide the enhancement process, which can lead to undesired artifacts and structural distortions. In this article, we propose a novel UMIE approach that avoids the above limitation of existing methods by directly encoding HQ cues into the LQ enhancement process in a variational fashion and thus model the UMIE task under the joint distribution between the LQ and HQ domains. Specifically, we extract features from an HQ image and explicitly insert the features, which are expected to encode HQ cues, into the enhancement network to guide the LQ enhancement with the variational normalization module. We train the enhancement network adversarially with a discriminator to ensure the generated HQ image falls into the HQ domain. We further propose a content-aware loss to guide the enhancement process with wavelet-based pixel-level and multiencoder-based feature-level constraints. Additionally, as a key motivation for performing image enhancement is to make the enhanced images serve better for downstream tasks, we propose a bi-level learning scheme to optimize the UMIE task and downstream tasks cooperatively, helping generate HQ images both visually appealing and favorable for downstream tasks. Experiments on three medical datasets verify that our method outperforms existing techniques in terms of both enhancement quality and downstream task performance. The code and the newly collected datasets are publicly available at https://github.com/ChunmingHe/HQG-Net.

Nonrigid registration of medical images is formulated usually as an optimization problem with the aim of seeking out the deformation field between a referential-moving image pair. During the past several years, advances have been achieved in the convolutional neural network (CNN)-based registration of images, whose performance was superior to most conventional methods. More lately, the long-range spatial correlations in images have been learned by incorporating an attention-based model into the transformer network. However, medical images often contain plural regions with structures that vary in size. The majority of the CNN- and transformer-based approaches adopt embedding of patches that are identical in size, disallowing representation of the inter-regional structural disparities within an image. Besides, it probably leads to the structural and semantical inconsistencies of objects as well. To address this issue, we put forward an innovative module called region-based structural relevance embedding (RSRE), which allows adaptive embedding of an image into unequally-sized structural regions based on the similarity of self-constructing latent graph instead of utilizing patches that are identical in size. Additionally, a transformer is integrated with the proposed module to serve as an adaptive region-based transformer (ART) for registering medical images nonrigidly. As demonstrated by the experimental outcomes, our ART is superior to the advanced nonrigid registration approaches in performance, whose Dice score is 0.734 on the LPBA40 dataset with 0.318% foldings for deformation field, and is 0.873 on the ADNI dataset with 0.331% foldings.

A key challenge of continual reinforcement learning (CRL) in dynamic environments is to promptly adapt the reinforcement learning (RL) agent's behavior as the environment changes over its lifetime while minimizing the catastrophic forgetting of the learned information. To address this challenge, in this article, we propose DaCoRL, that is, dynamics-adaptive continual RL. DaCoRL learns a context-conditioned policy using progressive contextualization, which incrementally clusters a stream of stationary tasks in the dynamic environment into a series of contexts and opts for an expandable multihead neural network to approximate the policy. Specifically, we define a set of tasks with similar dynamics as an environmental context and formalize context inference as a procedure of online Bayesian infinite Gaussian mixture clustering on environment features, resorting to online Bayesian inference to infer the posterior distribution over contexts. Under the assumption of a Chinese restaurant process (CRP) prior, this technique can accurately classify the current task as a previously seen context or instantiate a new context as needed without relying on any external indicator to signal environmental changes in advance. Furthermore, we employ an expandable multihead neural network whose output layer is synchronously expanded with the newly instantiated context and a knowledge distillation regularization term for retaining the performance on learned tasks. As a general framework that can be coupled with various deep RL algorithms, DaCoRL features consistent superiority over existing methods in terms of stability, overall performance, and generalization ability, as verified by extensive experiments on several robot navigation and MuJoCo locomotion tasks.

Multi-types of behaviors (e.g., clicking, carting, purchasing, etc.) widely exist in most real-world recommendation scenarios, which are beneficial to learn users’ multi-faceted preferences. As dependencies are explicitly exhibited by the multiple types of behaviors, effectively modeling complex behavior dependencies is crucial for multi-behavior prediction. The state-of-the-art multi-behavior models learn behavior dependencies indistinguishably with all historical interactions as input. However, different behaviors may reflect different aspects of user preference, which means that some irrelevant interactions may play as noises to the target behavior to be predicted. To address the aforementioned limitations, we introduce multi-interest learning to the multi-behavior recommendation. More specifically, we propose a novel Coarse-to-fine Knowledge-enhanced Multi-interest Learning (CKML) framework to learn shared and behavior-specific interests for different behaviors. CKML introduces two advanced modules, namely Coarse-grained Interest Extracting (CIE) and Fine-grained Behavioral Correlation (FBC) , which work jointly to capture fine-grained behavioral dependencies. CIE uses knowledge-aware information to extract initial representations of each interest. FBC incorporates a dynamic routing scheme to further assign each behavior among interests. Empirical results on three real-world datasets verify the effectiveness and efficiency of our model in exploiting multi-behavior data.

Source-free object detection (SFOD) aims to transfer a detector pre-trained on a label-rich source domain to an unlabeled target domain without seeing source data. While most existing SFOD methods generate pseudo labels via a source-pretrained model to guide training, these pseudo labels usually contain high noises due to heavy domain discrepancy. In order to obtain better pseudo supervisions, we divide the target domain into source-similar and source-dissimilar parts and align them in the feature space by adversarial learning.Specifically, we design a detection variance-based criterion to divide the target domain. This criterion is motivated by a finding that larger detection variances denote higher recall and larger similarity to the source domain. Then we incorporate an adversarial module into a mean teacher framework to drive the feature spaces of these two subsets indistinguishable. Extensive experiments on multiple cross-domain object detection datasets demonstrate that our proposed method consistently outperforms the compared SFOD methods. Our implementation is available at https://github.com/ChuQiaosong.

Blind Image Quality Assessment (BIQA) is a fundamental task in computer vision, which however remains unresolved due to the complex distortion conditions and diversified image contents. To confront this challenge, we in this paper propose a novel BIQA pipeline based on the Transformer architecture, which achieves an efficient quality-aware feature representation with much fewer data. More specifically, we consider the traditional fine-tuning in BIQA as an interpretation of the pre-trained model. In this way, we further introduce a Transformer decoder to refine the perceptual information of the CLS token from different perspectives. This enables our model to establish the quality-aware feature manifold efficiently while attaining a strong generalization capability. Meanwhile, inspired by the subjective evaluation behaviors of human, we introduce a novel attention panel mechanism, which improves the model performance and reduces the prediction uncertainty simultaneously. The proposed BIQA method maintains a light-weight design with only one layer of the decoder, yet extensive experiments on eight standard BIQA datasets (both synthetic and authentic) demonstrate its superior performance to the state-of-the-art BIQA methods, i.e., achieving the SRCC values of 0.875 (vs. 0.859 in LIVEC) and 0.980 (vs. 0.969 in LIVE). Checkpoints, logs and code will be available at https://github.com/narthchin/DEIQT.

Multi-behavior recommendation algorithms aim to leverage the multiplex interactions between users and items to learn users' latent preferences. Recent multi-behavior recommendation frameworks contain two steps: fusion and prediction. In the fusion step, advanced neural networks are used to model the hierarchical correlations between user behaviors. In the prediction step, multiple signals are utilized to jointly optimize the model with a multi-task learning (MTL) paradigm. However, recent approaches have not addressed the issue caused by imbalanced data distribution in the fusion step, resulting in the learned relationships being dominated by high-frequency behaviors. In the prediction step, the existing methods use a gate mechanism to directly aggregate expert information generated by coupling input, leading to negative information transfer. To tackle these issues, we propose a Parallel Knowledge Enhancement Framework (PKEF) for multi-behavior recommendation. Specifically, we enhance the hierarchical information propagation in the fusion step using parallel knowledge (PKF). Meanwhile, in the prediction step, we decouple the representations to generate expert information and introduce a projection mechanism during aggregation to eliminate gradient conflicts and alleviate negative transfer (PME). We conduct comprehensive experiments on three real-world datasets to validate the effectiveness of our model. The results further demonstrate the rationality and effectiveness of the designed PKF and PME modules. The source code and datasets are available at https://github.com/MC-CV/PKEF.

Query classification plays a crucial role in e-commerce, where the goal is to assign user queries to appropriate categories within a hierarchical product category taxonomy. However, existing methods rely on a limited number of words from the category description and often neglect the hierarchical structure of the category tree, resulting in suboptimal category representations. To overcome these limitations, we propose a novel approach named hierarchical contrastive learning framework for query classification (HCL4QC), which leverages the hierarchical category tree structure to improve the performance of query classification. Specifically, HCL4QC is designed as a plugin module that consists of two innovative losses, namely local hierarchical contrastive loss (LHCL) and global hierarchical contrastive loss (GHCL). LHCL adjusts representations of categories according to their positional relationship in the hierarchical tree, while GHCL ensures the semantic consistency between the parent category and its child categories. Our proposed method can be adapted to any query classification tasks that involve a hierarchical category structure. We conduct experiments on two real-world datasets to demonstrate the superiority of our hierarchical contrastive learning. The results demonstrate significant improvements in the query classification task, particularly for long-tail categories with sparse supervised information.

Camouflaged object detection (COD) aims to address the tough issue of identifying camouflaged objects visually blended into the surrounding backgrounds. COD is a challenging task due to the intrinsic similarity of camouflaged objects with the background, as well as their ambiguous boundaries. Existing approaches to this problem have developed various techniques to mimic the human visual system. Albeit effective in many cases, these methods still struggle when camouflaged objects are so deceptive to the vision system. In this paper, we propose the FEature Decomposition and Edge Reconstruction (FEDER) model for COD. The FEDER model addresses the intrinsic similarity of foreground and background by decomposing the features into different frequency bands using learnable wavelets. It then focuses on the most informative bands to mine subtle cues that differentiate foreground and background. To achieve this, a frequency attention module and a guidance-based feature aggregation module are developed. To combat the ambiguous boundary problem, we propose to learn an auxiliary edge reconstruction task alongside the COD task. We design an ordinary differential equation-inspired edge reconstruction module that generates exact edges. By learning the auxiliary task in conjunction with the COD task, the FEDER model can generate precise prediction maps with accurate object boundaries. Experiments show that our FEDER model significantly outperforms state-of-the-art methods with cheaper computational and memory costs. The code will be available at https://github.com/ChunmingHe/FEDER.

With the continuously thriving popularity around the world, fitness activity analytic has become an emerging research topic in computer vision. While a variety of new tasks and algorithms have been proposed recently, there are growing hunger for data resources involved in high-quality data, fine-grained labels, and diverse environments. In this paper, we present FLAG3D, a large-scale 3D fitness activity dataset with language instruction containing 180K sequences of 60 categories. FLAG3D features the following three aspects: 1) accurate and dense 3D human pose captured from advanced MoCap system to handle the complex activity and large movement, 2) detailed and professional language instruction to describe how to perform a specific activity, 3) versatile video resources from a high-tech MoCap system, rendering software, and cost-effective smartphones in natural environments. Extensive experiments and in-depth analysis show that FLAG3D contributes great research value for various challenges, such as cross-domain human action recognition, dynamic human mesh recovery, and language-guided human action generation. Our dataset and source code are publicly available at https://andytang15.github.io/FLAG3D.

Equipped with the trained environmental dynamics, model-based offline reinforcement learning (RL) algorithms can often successfully learn good policies from fixed-sized datasets, even some datasets with poor quality. Unfortunately, however, it can not be guaranteed that the generated samples from the trained dynamics model are reliable (e.g., some synthetic samples may lie outside of the support region of the static dataset). To address this issue, we propose Trajectory Truncation with Uncertainty (TATU), which adaptively truncates the synthetic trajectory if the accumulated uncertainty along the trajectory is too large. We theoretically show the performance bound of TATU to justify its benefits. To empirically show the advantages of TATU, we first combine it with two classical model-based offline RL algorithms, MOPO and COMBO. Furthermore, we integrate TATU with several off-the-shelf model-free offline RL algorithms, e.g., BCQ. Experimental results on the D4RL benchmark show that TATU significantly improves their performance, often by a large margin. Code is available here.

Interactive devices with touch screen have become commonly used in various aspects of daily life, which raises the demand for high production quality of touch screen glass. While it is desirable to develop effective defect detection technologies to optimize the automatic touch screen production lines, the development of these technologies suffers from the lack of publicly available datasets. To address this issue, we in this paper propose a dedicated touch screen glass defect dataset which includes seven types of defects and consists of 2504 images captured in various scenarios. All data are captured with professional acquisition equipment on the fixed workstation. Additionally, we benchmark the CNN- and Transformer-based object detection frameworks on the proposed dataset to demonstrate the challenges of defect detection on high-resolution images. Dataset and related code will be available at https://github.com/VincentHancoder/SSGD.

Semi-Supervised Domain Adaptation (SSDA) is a recently emerging research topic that extends from the widely-investigated Unsupervised Domain Adaptation (UDA) by further having a few target samples labeled, i.e., the model is trained with labeled source samples, unlabeled target samples as well as a few labeled target samples. Compared with UDA, the key to SSDA lies how to most effectively utilize the few labeled target samples. Existing SSDA approaches simply merge the few precious labeled target samples into vast labeled source samples or further align them, which dilutes the value of labeled target samples and thus still obtains a biased model. To remedy this, in this paper, we propose to decouple SSDA as an UDA problem and a semi-supervised learning problem where we first learn an UDA model using labeled source and unlabeled target samples and then adapt the learned UDA model in a semi-supervised way using labeled and unlabeled target samples. By utilizing the labeled source samples and target samples separately, the bias problem can be well mitigated. We further propose a consistency learning based mean teacher model to effectively adapt the learned UDA model using labeled and unlabeled target samples. Experiments show our approach outperforms existing methods.

In this paper, we propose SemanticAC, a semantics-assisted framework for Audio Classification to better leverage the semantic information. Unlike conventional audio classification methods that treat class labels as discrete vectors, we employ a language model to extract abundant semantics from labels and optimize the semantic consistency between audio signals and their labels. We verify that simple textual information from labels and advanced pretraining models enable more abundant semantic supervision for better performance. Specifically, we design a text encoder to capture the semantic information from the text extension of labels. Then we map the audio signals to align with the semantics of corresponding class labels via an audio encoder and a similarity calculation module so as to enforce the semantic consistency. Extensive experiments on two audio datasets, ESC-50 and US8K demonstrate that our proposed method consistently outperforms the compared audio classification methods.

Heterogeneous image fusion (HIF) techniques aim to enhance image quality by merging complementary information from images captured by different sensors. Among these algorithms, deep unfolding network (DUN)-based methods achieve promising performance but still suffer from two issues: they lack a degradation-resistant-oriented fusion model and struggle to adequately consider the structural properties of DUNs, making them vulnerable to degradation scenarios. In this paper, we propose a Degradation-Resistant Unfolding Network (DeRUN) for the HIF task to generate high-quality fused images even in degradation scenarios. Specifically, we introduce a novel HIF model for degradation resistance and derive its optimization procedures. Then, we incorporate the optimization unfolding process into the proposed DeRUN for end-to-end training. To ensure the robustness and efficiency of DeRUN, we employ a joint constraint strategy and a lightweight partial weight sharing module. To train DeRUN, we further propose a gradient direction-based entropy loss with powerful texture representation capacity. Extensive experiments show that DeRUN significantly outperforms existing methods on four HIF tasks, as well as downstream applications, with cheaper computational and memory costs.

Generating full-body and multi-genre dance sequences from given music is a challenging task, due to the limitations of existing datasets and the inherent complexity of the fine-grained hand motion and dance genres. To address these problems, we propose FineDance, which contains 14.6 hours of music-dance paired data, with fine-grained hand motions, fine-grained genres (22 dance genres), and accurate posture. To the best of our knowledge, FineDance is the largest music-dance paired dataset with the most dance genres. Additionally, to address monotonous and unnatural hand movements existing in previous methods, we propose a full-body dance generation network, which utilizes the diverse generation capabilities of the diffusion model to solve monotonous problems, and use expert nets to solve unreal problems. To further enhance the genre-matching and long-term stability of generated dances, we propose a Genre&Coherent aware Retrieval Module. Besides, we propose a novel metric named Genre Matching Score to evaluate the genre-matching degree between dance and music. Quantitative and qualitative experiments demonstrate the quality of FineDance, and the state-of-the-art performance of FineNet. The FineDance Dataset and more qualitative samples can be found at website.

Box-supervised instance segmentation has gained much attention as it requires only simple box annotations instead of costly mask or polygon annotations. However, existing box-supervised instance segmentation models mainly focus on mask-based frameworks. We propose a new end-to-end training technique, termed BoxSnake, to achieve effective polygonal instance segmentation using only box annotations for the first time. Our method consists of two loss functions: (1) a point-based unary loss that constrains the bounding box of predicted polygons to achieve coarse-grained segmentation; and (2) a distance-aware pairwise loss that encourages the predicted polygons to fit the object boundaries. Compared with the mask-based weakly-supervised methods, BoxSnake further reduces the performance gap between the predicted segmentation and the bounding box, and shows significant superiority on the Cityscapes dataset. The source code has been available at https://github.com/Yangr116/BoxSnake.

In this paper, we propose a context-aware inpainter-refiner (CAIR) framework for skeleton-based human motion completion, which aims to restore the damaged area of a given corrupted sequence in spatio-temporal domain. Conventional methods usually utilize convolutional neural networks to inpaint the corrupted sequence treated as a image. In comparison, we devise a two-stage "inpainting-refining" framework and further design a context-aware graph convolutional module (CA-GCM), which employs graph convolutional operation and mask-aware residual connection to update values of corrupted area. In this way, our CAIR is capable of mapping and associating intact and corrupted joints implicitly, and thus optimizes the result generated by our inpainting network effectively. Extensive experimental results on NTU60 and Human3.6M datasets demonstrate that our method outperforms our own baseline and previous methods.

Exoskeleton robots provide assistive forces to suit the human subject via physical human-robot interaction. During the closely-coupled interaction, a mismatch between the wearer and the robot may result in physical conflict, which could affect assistance efficiency or even compromise safety. Therefore, such conflicts should be accurately detected and then properly relaxed by adjusting the robot's action. This paper proposes a new learning scheme to detect physical conflicts between humans and robots. The constructed learning network receives multi-modal information from proprioceptive sensors and then outputs the anomaly score to specify the physical conflict, which score is further used to continuously adjust the robot impedance to ensure a safe and efficient interaction. Such a formulation allows the robot to explore the semantic information during the interaction (e.g., gait phases, imbalance, human fatigue) and hence react properly to the physical conflict. Experimental results and comparative studies on a lower-limb exoskeleton robot are presented to illustrate that the proposed learning scheme can deal with physical conflicts in a faster and more accurate manner.

Visual anomaly detection methods with localization are critically important for industrial manufacturing quality control. Because of the rarity of anomalies and the irregular variation of anomaly patterns, unsupervised methods have been widely explored. For the anomaly detection and localization tasks, the image reconstruction-based approaches have shown competitive performance. However, reconstruction results of those methods are coarse and visually blurred, which leads to a high rate of false detection and false pixel-level localization. To address those issues, we propose a framework called Dual-Path Re-construction Guided Segmentation Network (DRGS-Net), which determines the abnormal regions by segmenting the anomaly image with its reconstructed result as the reference template. DRGS-Net mainly consists of a novel dual-path reconstruction sub-network and a specifically designed anomaly segmentation sub-network. They are jointly trained end-to-end, with the former fusing texture repair and image reconstruction information to obtain fine-grained reconstruction results and the latter learning a decision boundary between normal and anomalous regions based on reconstruction results. On the standard benchmark dataset MVTecAD and an additional dataset DAGM, DRGS-Net shows competitive performance in image-level detection and achieves outstanding improvement in pixel-level localization. Further experiments with the few samples setting demonstrate that DRGS-Net retains strong performance with only few anomaly-free training images.

Various types of user behaviors are recorded in most real-world recommendation scenarios. To fully utilize the multi-behavior information, the exploration of multiplex interaction among them is essential. Many multi-task learning based multi-behavior methods are proposed recently to use multiple types of supervision signals and perform information transfer among them. Despite the great successes, these methods fail to design prediction tasks comprehensively, leading to insufficient utilization of multi-behavior correlative information. Besides, these methods are either based on the weighting of expert information extracted from the coupled input or modeling of information transfer between multiple behavior levels through task-specific extractors, which are usually accompanied by negative transfer phenomenon1. To address the above problems, we propose a multi-behavior recommendation framework, called Hierarchical Projection Enhanced Multi-behavior Recommendation (HPMR). The key module, Projection-based Transfer Network (PTN), uses the projection mechanism to "explicitly" model the correlations of upstream and downstream behaviors, refines the upstream behavior representations, and fully uses the refined representations to enhance the learning of downstream tasks. Offline experiments on public and industrial datasets and online A/B test further verify the effectiveness of HPMR in modeling the associations from upstream to downstream and alleviating the negative transfer. The source code and datasets are available at https://github.com/MC-CV/HPMR.

Graph neural networks, which typically exchange information between local neighbors, often struggle to capture long-range interactions (LRIs) within the graph. Building a graph hierarchy via graph pooling methods is a promising approach to address this challenge; however, hierarchical information propagation cannot entirely take over the role of local information aggregation. To balance locality and hierarchy, we integrate the local and hierarchical structures, represented by intra- and inter-graphs respectively, of a multi-scale graph hierarchy into a single mega graph. Our proposed MeGraph model consists of multiple layers alternating between local and hierarchical information aggregation on the mega graph. Each layer first performs local-aware message-passing on graphs of varied scales via the intra-graph edges, then fuses information across the entire hierarchy along the bidirectional pathways formed by inter-graph edges. By repeating this fusion process, local and hierarchical information could intertwine and complement each other. To evaluate our model, we establish a new Graph Theory Benchmark designed to assess LRI capture ability, in which MeGraph demonstrates dominant performance. Furthermore, MeGraph exhibits superior or equivalent performance to state-of-the-art models on the Long Range Graph Benchmark. The experimental results on commonly adopted real-world datasets further demonstrate the broad applicability of MeGraph.

Weakly-Supervised Concealed Object Segmentation (WSCOS) aims to segment objects well blended with surrounding environments using sparsely-annotated data for model training. It remains a challenging task since (1) it is hard to distinguish concealed objects from the background due to the intrinsic similarity and (2) the sparsely-annotated training data only provide weak supervision for model learning. In this paper, we propose a new WSCOS method to address these two challenges. To tackle the intrinsic similarity challenge, we design a multi-scale feature grouping module that first groups features at different granularities and then aggregates these grouping results. By grouping similar features together, it encourages segmentation coherence, helping obtain complete segmentation results for both single and multiple-object images. For the weak supervision challenge, we utilize the recently-proposed vision foundation model, ``Segment Anything Model (SAM)'', and use the provided sparse annotations as prompts to generate segmentation masks, which are used to train the model. To alleviate the impact of low-quality segmentation masks, we further propose a series of strategies, including multi-augmentation result ensemble, entropy-based pixel-level weighting, and entropy-based image-level selection. These strategies help provide more reliable supervision to train the segmentation model. We verify the effectiveness of our method on various WSCOS tasks, and experiments demonstrate that our method achieves state-of-the-art performance on these tasks.

This paper studies referring video object segmentation (RVOS) by boosting video-level visual-linguistic alignment. Recent approaches model the RVOS task as a sequence prediction problem and perform multi-modal interaction as well as segmentation for each frame separately. However, the lack of a global view of video content leads to difficulties in effectively utilizing inter-frame relationships and understanding textual descriptions of object temporal variations. To address this issue, we propose Semantic-assisted Object Cluster (SOC), which aggregates video content and textual guidance for unified temporal modeling and cross-modal alignment. By associating a group of frame-level object embeddings with language tokens, SOC facilitates joint space learning across modalities and time steps. Moreover, we present multi-modal contrastive supervision to help construct well-aligned joint space at the video level. We conduct extensive experiments on popular RVOS benchmarks, and our method outperforms state-of-the-art competitors on all benchmarks by a remarkable margin. Besides, the emphasis on temporal coherence enhances the segmentation stability and adaptability of our method in processing text expressions with temporal variations. Code is available at https://github.com/RobertLuo1/NeurIPS2023_SOC.

This paper aims to efficiently enable Large Language Models (LLMs) to use multi-modal tools. The advanced proprietary LLMs, such as ChatGPT and GPT-4, have shown great potential for tool usage through sophisticated prompt engineering. Nevertheless, these models typically rely on prohibitive computational costs and publicly inaccessible data. To address these challenges, we propose the GPT4Tools based on self-instruct to enable open-source LLMs, such as LLaMA and OPT, to use tools. It generates an instruction-following dataset by prompting an advanced teacher with various multi-modal contexts. By using the Low-Rank Adaptation (LoRA) optimization, our approach facilitates the open-source LLMs to solve a range of visual problems, including visual comprehension and image generation. Moreover, we provide a benchmark to evaluate the ability of LLMs to use tools, which is performed in both zero-shot and fine-tuning ways. Extensive experiments demonstrate the effectiveness of our method on various language models, which not only significantly improves the accuracy of invoking seen tools, but also enables the zero-shot capacity for unseen tools.

Multi-types of user behavior data (e.g., clicking, adding to cart, and purchasing) are recorded in most real-world recommendation scenarios, which can help to learn users’ multi-faceted preferences. However, it is challenging to explore multi-behavior data due to the unbalanced data distribution and sparse target behavior, which lead to the inadequate modeling of high-order relations when treating multi-behavior data “as features” and gradient conflict in multi-task learning when treating multi-behavior data “as labels”. In this paper, we propose CIGF, a Compressed Interaction Graph based Framework, to overcome the above limitations. Specifically, we design a novel Compressed Interaction Graph Convolution Network (CIGCN) to model instance-level high-order relations explicitly. To alleviate the potential gradient conflict when treating multi-behavior data “as labels”, we propose a Multi-Expert with Separate Input (MESI) network with separate input on the top of CIGCN for multi-task learning. Comprehensive experiments on three large-scale real-world datasets demonstrate the superiority of CIGF.

Video captioning is a challenging task of automatically generating natural and meaningful textual descriptions given some context videos. The state-of-the-art methods aggregate the spatial-wise information in the video encoder at the early stage, which has two drawbacks: 1) Early aggregation in the encoder can cause considerable spatial details missing, which may consequently lead to incorrect word choices in the following text encoder. 2) The spatial attention learned in the video encoder may not be compelling enough without text guidance. To solve these problems, we propose a Stay-in-Grid video CAPtioning method SGCAP, which makes full use of the grid-level spatial features and consists of a Bilinear Sequential Attention Encoder (BSAE) and a Cross-modal Sequential Attention Decoder (CSAD). The former explores and retains fully grid-level discriminative representations in the video encoder, while the latter performs the late spatial aggregation in the decoder to attend to the most relevant regions with the supervision of the input words. Experimental results demonstrate the effectiveness of our method on three public datasets, showing its superior performance over multiple state-of-the-art video captioning models. Source codes and the pre-trained models will be made available to the public.

In this paper, we propose a discrepancy-aware meta-learning approach for zero-shot face manipulation detection, which aims to learn a discriminative model maximizing the generalization to unseen face manipulation attacks with the guidance of the discrepancy map. Unlike existing face manipulation detection methods that usually present algorithmic solutions to the known face manipulation attacks, where the same types of attacks are used to train and test the models, we define the detection of face manipulation as a zero-shot problem. We formulate the learning of the model as a meta-learning process and generate zero-shot face manipulation tasks for the model to learn the meta-knowledge shared by diversified attacks. We utilize the discrepancy map to keep the model focused on generalized optimization directions during the meta-learning process. We further incorporate a center loss to better guide the model to explore more effective meta-knowledge. Experimental results on the widely used face manipulation datasets demonstrate that our proposed approach achieves very competitive performance under the zero-shot setting.

How to obtain good value estimation is a critical problem in Reinforcement Learning (RL). Current value estimation methods in continuous control, such as DDPG and TD3, suffer from unnecessary over- or under- estimation. In this paper, we explore the potential of double actors, which has been neglected for a long time, for better value estimation in the continuous setting. First, we interestingly find that double actors improve the exploration ability of the agent. Next, we uncover the bias alleviation property of double actors in handling overestimation with single critic, and underestimation with double critics respectively. Finally, to mitigate the potentially pessimistic value estimate in double critics, we propose to regularize the critics under double actors architecture. Together, we present Double Actors Regularized Critics (DARC) algorithm. Extensive experiments on challenging continuous control benchmarks, MuJoCo and PyBullet, show that DARC significantly outperforms current baselines with higher average return and better sample efficiency.

Sequential recommendation (SR) plays an important role in personalized recommender systems because it captures dynamic and diverse preferences from users' real-time increasing behaviors. Unlike the standard autoregressive training strategy, future data (also available during training) has been used to facilitate model training as it provides richer signals about users' current interests and can be used to improve the recommendation quality. However, existing methods suffer from a severe training-inference gap, i.e., both past and future contexts are modeled by the same encoder when training, while only historical behaviors are available during inference. This discrepancy leads to potential performance degradation. To alleviate the training-inference gap, we propose a new framework DualRec, which achieves past-future disentanglement and past-future mutual enhancement by a novel dual network. Specifically, a dual network structure is exploited to model the past and future context separately.And a bi-directional knowledge transferring mechanism enhances the knowledge learnt by the dual network. Extensive experiments on four real-world datasets demonstrate the superiority of our approach over baseline methods. Besides, we demonstrate the compatibility of DualRec by instantiating using different backbones. Further empirical analysis verifies the high utility of modeling future contexts under our DualRec framework.

Anomaly Detection (AD) aims to find defective patterns or abnormal samples among data, and has been a hot research topic due to various real-world applications. While various AD methods have been proposed, most of them assume the availability of a clean (anomaly-free) training set, which however may be hard to guarantee in many real-world industry applications. This motivates us to investigate Unsupervised Anomaly Detection (UAD) in which the training set includes both normal and abnormal samples. In this paper, we address the UAD problem by proposing a Self-Training and Knowledge Distillation (STKD) model. STKD combats anomalies in the training set by iteratively alternating between excluding samples of high anomaly probabilities and training the model with the purified training set. Despite that the model is trained with a cleaner training set, the inevitably existing anomalies may still cause negative impact. STKD alleviates this by regularizing the model to respond similarly to a teacher model which has not been trained with noisy data. Experiments show that STKD consistently produces more robust performance with different levels of anomalies.

Laser-driven micro-tools are formulated by treating highly-focused laser beams as actuators, to control the tool's motion to contact then manipulate a micro object, which allows it to manipulate opaque micro objects, or large cells without causing photodamage. However, most existing laser-driven tools are limited to relatively simple tasks, such as moving and caging, and cannot carry out in-hand dexterous tasks. This is mainly because in-hand manipulation involves continuously coordinating multiple laser beams, micro-tools, and the object itself, which has high degrees of freedom (DoF) and poses up challenge for planner and controller design. This paper presents a new hierarchical formulation for the grasping and manipulation of micro objects using multiple laser-driven micro-tools. In hardware, multiple laser-driven tools are assembled to act as a robotic hand to carry out in-hand tasks (e.g., rotating); in software, a hierarchical scheme is developed to shrunken the action space and coordinate the motion of multiple tools, subject to both the parametric uncertainty in the tool and the unknown dynamic model of the object. Such a formulation provides potential for achieving robotic in-hand manipulation at a micro scale. The performance of the proposed system is validated in simulation studies under different scenarios.

The style-based generator architectures (e.g. StyleGAN v1, v2) largely promote the controllability and explainability of Generative Adversarial Networks (GANs). Many researchers have applied the pretrained style-based generators to image manipulation and video editing by exploring the correlation between linear interpolation in the latent space and semantic transformation in the synthesized image manifold. However, most previous studies focused on manipulating separate discrete attributes, which is insufficient to animate a still image to generate videos with complex and diverse poses and expressions. In this work, we devise a dual distillation strategy (D2Animator) for generating animated high-resolution face videos conditioned on identities and poses from different images. Specifically, we first introduce a Clustering-based Distiller (CluDistiller) to distill diverse interpolation directions in the latent space, and synthesize identity-consistent faces with various poses and expressions, such as blinking, frowning, looking up/down, etc. Then we propose an Augmentation-based Distiller (AugDistiller) that learns to encode arbitrary face deformation into a combination of interpolation directions via training on augmentation samples synthesized by CluDistiller. Through assembling the two distillation methods, D2Animator can generate high-resolution face animation videos without training on video sequences. Extensive experiments on self-driving, cross-identity and sequence-driving tasks demonstrate the superiority of the proposed D2Animator over existing StyleGAN manipulation and face animation methods in both generation quality and animation fidelity.

Action recognition has been traditionally treated as a high-level video classification problem. However, such a manner lacks the detailed and semantic understanding of body movement, which is the critical knowledge to explain and infer complex human actions. To fill this gap, we propose to summarize a novel visual knowledge graph from over 15M detailed human annotations, for describing action as the distinct composition of body parts, part movements and interactive objects in videos. Based on it, we design a generic multi-modal Action Knowledge Understanding (AKU) framework, which can progressively infer human actions from body part movements in the videos, with assistance of visual-driven semantic knowledge mining. Finally, we validate AKU on the recent Kinetics-TPS benchmark, which contains body part parsing annotations for detailed understanding of human action in videos. The results show that, our AKU significantly boosts various video backbones with explainable action knowledge in both supervised and few shot settings, and outperforms the recent knowledge-based action recognition framework, e.g., our AKU achieves 83.9% accuracy on Kinetics-TPS while PaStaNet achieves 63.8% accuracy under the same backbone. The codes and models will be released at https://github.com/mayuelala/AKU.

This paper presents a language-powered paradigm for ordinal regression. Existing methods usually treat each rank as a category and employ a set of weights to learn these concepts. These methods are easy to overfit and usually attain unsatisfactory performance as the learned concepts are mainly derived from the training set. Recent large pre-trained vision-language models like CLIP have shown impressive performance on various visual tasks. In this paper, we propose to learn the rank concepts from the rich semantic CLIP latent space. Specifically, we reformulate this task as an image-language matching problem with a contrastive objective, which regards labels as text and obtains a language prototype from a text encoder for each rank. While prompt engineering for CLIP is extremely time-consuming, we propose OrdinalCLIP, a differentiable prompting method for adapting CLIP for ordinal regression. OrdinalCLIP consists of learnable context tokens and learnable rank embeddings. The learnable rank embeddings are constructed by explicitly modeling numerical continuity, resulting in well-ordered, compact language prototypes in the CLIP space. Once learned, we can only save the language prototypes and discard the huge language model, resulting in zero additional computational overhead compared with the linear head counterpart. Experimental results show that our paradigm achieves competitive performance in general ordinal regression tasks, and gains improvements in few-shot and distribution shift settings for age estimation. The code is available at https://github.com/xk-huang/OrdinalCLIP.

The learned policy of model-free offline reinforcement learning (RL) methods is often constrained to stay within the support of datasets to avoid possible dangerous out-of-distribution actions or states, making it challenging to handle out-of-support region. Model-based RL methods offer a richer dataset and benefit generalization by generating imaginary trajectories with either trained forward or reverse dynamics model. However, the imagined transitions may be inaccurate, thus downgrading the performance of the underlying offline RL method. In this paper, we propose to augment the offline dataset by using trained bidirectional dynamics models and rollout policies with double check. We introduce conservatism by trusting samples that the forward model and backward model agree on. Our method, confidence-aware bidirectional offline model-based imagination, generates reliable samples and can be combined with any model-free offline RL method. Experimental results on the D4RL benchmarks demonstrate that our method significantly boosts the performance of existing model-free offline RL algorithms and achieves competitive or better scores against baseline methods.

Offline reinforcement learning (RL) defines the task of learning from a static logged dataset without continually interacting with the environment. The distribution shift between the learned policy and the behavior policy makes it necessary for the value function to stay conservative such that out-of-distribution (OOD) actions will not be severely overestimated. However, existing approaches, penalizing the unseen actions or regularizing with the behavior policy, are too pessimistic, which suppresses the generalization of the value function and hinders the performance improvement. This paper explores mild but enough conservatism for offline learning while not harming generalization. We propose Mildly Conservative Q-learning (MCQ), where OOD actions are actively trained by assigning them proper pseudo Q values. We theoretically show that MCQ induces a policy that behaves at least as well as the behavior policy and no erroneous overestimation will occur for OOD actions. Experimental results on the D4RL benchmarks demonstrate that MCQ achieves remarkable performance compared with prior work. Furthermore, MCQ shows superior generalization ability when transferring from offline to online, and significantly outperforms baselines. Our code is publicly available at https://github.com/dmksjfl/MCQ.

Following type-1 fractional fuzzy inference systems presented recently as the new generation of fuzzy inference systems, interval type-2 fractional fuzzy inference systems (IT2FFISs) as a leap further ahead in the evolution of fuzzy inference systems (FISs) are introduced in this article. The IT2FFISs, which are outlined in this article, add to the armamentarium of FISs some particular concepts such as interval type-2 fractional membership functions, type-2 fractional translation rule, type-2 fracture index, the concept of switching, the entanglement, the degeneracy concept, and so forth. An IT2FFIS exploits not only the tolerance for the uncertainty in the interpretation of the meaning of a word, but also the relevance between the quality and quantity levels of the given information to infer an answer to an inference query. The IT2FFISs make an increase in machine intelligence quotient possible by an increase in the range of FISs order rather than their type. Moreover, the synergy of the concepts coming with various modes of IT2FFISs such as the aggressive mode opens a gate to a space of fuzzy systems outputs which used to be indiscoverable. Furthermore, it is demonstrated that as the type-2 fracture index approaches zero, the space of IT2FFISs outputs contracts and eventually it coincides the space of IT2FISs output when the fracture index is equal to zero. It is also proved that, provided a particular order of the IT2FFIS is taken into account, independent of the problem in question, a typical IT2FIS never leads to results which are more satisfactory than those obtained by the IT2FFIS corresponding to the typical IT2FIS.

In this paper, we propose a Structure-adaptive Neighborhood Preserving Hashing (SNPH) method for unsupervised scalable video search. Unlike most existing hashing methods which equally encode an entire video into a binary feature vector, we propose a neighborhood attention mechanism which encodes the neighborhood-relevant content of a video to better preserve the neighborhood relationships among videos. Motivated by the fact that a video usually contains multiple shots and each shot depicts a different activity, we further develop a structure-adaptive encoder to model the hierarchical structure of the video. Specifically, the encoder adaptively divides each video into multiple segments via detecting temporal boundaries across frames and encodes these segments as a compact binary vector to capture rich structural information. We integrate the neighborhood attention mechanism into the structure-adaptive encoder to learn hash functions that jointly preserve the neighborhood relationships among videos and exploit the hierarchical structure in a video. Experimental results on three widely used benchmark datasets show that our proposed method consistently outperforms state-of-the-art unsupervised video hashing methods.

Currently, fingerprint authentication systems in mobile devices, which have limited-size fingerprint sensors, are mainly based on partial fingerprint matching algorithms. To cover all areas of the finger, the system usually collects multiple partially overlapping partial fingerprints during the enrollment. Existing recognition methods either perform score-level fusion after single-to-single matching, or perform single-to-single matching after image-level mosaicking. However, these two-stage methods have the risk of discarding some real information or introducing some fake information. In this paper, we define this “query2set” task and propose a novel single-to-multiple partial fingerprint recognition method based on atttention mechanism. Our end-to-end deep model can adaptively extract and fuse appropriate features from a set of fingerprints for matching based on the input query fingerprint. Experiments indicate that our method outperforms several state-of-the-art single-to-single approaches and provides a new insight of fingerprint recognition on mobile devices.

In this paper, we propose a salience-aware face presentation attack detection (SAFPAD) approach, which takes advantage of deep reinforcement learning to exploit the salient local part information in face images. Most existing deep face presentation attack detection approaches extract features from the entire image or several fixed regions. However, the discriminative information beneficial for presentation attack detection is unevenly distributed in the image due to the illumination and presentation attack instrument variation, so treating all regions equally fails to highlight the most discriminative information which is important for more accurate and robust face presentation attack detection. To address this, we propose to identify the discriminative salient parts using deep reinforcement learning and focus on them to alleviate the adverse effects of redundant information in the face images. We fuse the high-level features and the local features which guide the policy network to exploit discriminative patches and assist the classification network to predict more accurate results. We jointly train the SAFPAD model with deep reinforcement learning to generate salient locations. Extensive experiments on five public datasets demonstrate that our approach achieves very competitive performance due to the concentrated employment of salient local information.

Automated radiographic report generation is challenging in at least two aspects. First, medical images are very similar to each other and the visual differences of clinic importance are often fine-grained. Second, the disease-related words may be submerged by many similar sentences describing the common content of the images, causing the abnormal to be misinterpreted as the normal in the worst case. To tackle these challenges, this paper proposes a pure transformer-based framework to jointly enforce better visual-textual alignment, multi-label diagnostic classification, and word importance weighting, to facilitate report generation. To the best of our knowledge, this is the first pure transformer-based framework for medical report generation, which enjoys the capacity of transformer in learning long range dependencies for both image regions and sentence words. Specifically, for the first challenge, we design a novel mechanism to embed an auxiliary image-text matching objective into the transformer's encoder-decoder structure, so that better correlated image and text features could be learned to help a report to discriminate similar images. For the second challenge, we integrate an additional multi-label classification task into our framework to guide the model in making correct diagnostic predictions. Also, a term-weighting scheme is proposed to reflect the importance of words for training so that our model would not miss key discriminative information. Our work achieves promising performance over the state-of-the-arts on two benchmark datasets, including the largest dataset MIMIC-CXR.

Most existing unsupervised video hashing methods are built on unidirectional models with less reliable training objectives, which underuse the correlations among frames and the similarity structure between videos. To enable efficient scalable video retrieval, we propose a self-supervised video Hashing method based on Bidirectional Transformers (BTH). Based on the encoder-decoder structure of transformers, we design a visual cloze task to fully exploit the bidirectional correlations between frames. To unveil the similarity structure between unlabeled video data, we further develop a similarity reconstruction task by establishing reliable and effective similarity connections in the video space. Furthermore, we develop a cluster assignment task to exploit the structural statistics of the whole dataset such that more discriminative binary codes can be learned. Extensive experiments implemented on three public benchmark datasets, FCVID, ActivityNet and YFCC, demonstrate the superiority of our proposed approach.

Automated radiographic report generation is a challenging task since it requires to generate paragraphs describing fine-grained visual differences of cases, especially for those between the diseased and the healthy. Existing image captioning methods commonly target at generic images, and lack mechanism to meet this requirement. To bridge this gap, in this paper, we propose a self-boosting framework that improves radiographic report generation based on the cooperation of the main task of report generation and an auxiliary task of image-text matching. The two tasks are built as the two branches of a network model and influence each other in a cooperative way. On one hand, the image-text matching branch helps to learn highly text-correlated visual features for the report generation branch to output high quality reports. On the other hand, the improved reports produced by the report generation branch provide additional harder samples for the image-text matching branch and enforce the latter to improve itself by learning better visual and text feature representations. This, in turn, helps improve the report generation branch again. These two branches are jointly trained to help improve each other iteratively and progressively, so that the whole model is self-boosted without requiring external resources. Experimental results demonstrate the effectiveness of our method on two public datasets, showing its superior performance over multiple state-of-the-art image captioning and medical report generation methods.

Deep neural networks (DNNs) have been widely used in facial manipulation. Existing methods focus on training deeper networks in indirect supervision ways (e.g., feature constraint), or in unsupervised ways (e.g., cycle-consistency loss) due to the lack of ground-truth face images for manipulated outputs. However, such methods can not synthesize realistic face images well and suffer from very high training overhead. To address this issue, we propose a novel Feature Disentanglement and Reintegraion network (FDRNet), which employs ground-truth images as informative supervision and dynamically adapts the fusion of informative features of the ground-truth images effectively and efficiently. FDRNet consists of a Feature Disentanglement (FD) Network and a Feature Reintegration (FR) Network, which encodes informative disentangled representations from the ground-truth images and fuses the disentangled representations to reconstruct the face images. By learning disentangled representations, our method can generate plausible faces conditioned on both landmarks and identities, which can be used for a variety of face manipulation tasks. Experiments on the CelebA-HQ and FFHQ datasets are conducted to demonstrate the superiority of our method over state-of-the-art methods in terms of effectiveness and efficiency.

Multimodal image registration (MIR) is a fundamental procedure in many image-guided therapies. Recently, unsupervised learning-based methods have demonstrated promising performance over accuracy and efficiency in deformable image registration. However, the estimated deformation fields of the existing methods fully rely on the to-be-registered image pair. It is difficult for the networks to be aware of the mismatched boundaries, resulting in unsatisfactory organ boundary alignment. In this paper, we propose a novel multimodal registration framework, which elegantly leverages the deformation fields estimated from both: (i) the original to-be-registered image pair, (ii) their corresponding gradient intensity maps, and adaptively fuses them with the proposed gated fusion module. With the help of auxiliary gradient-space guidance, the network can concentrate more on the spatial relationship of the organ boundary. Experimental results on two clinically acquired CT-MRI datasets demonstrate the effectiveness of our proposed approach.

In this paper, we propose a Frequency-Aware Spatiotemporal Transformer (FAST) for video inpainting detection, which aims to simultaneously mine the traces of video in-painting from spatial, temporal, and frequency domains. Unlike existing deep video inpainting detection methods that usually rely on hand-designed attention modules and memory mechanism, our proposed FAST have innate global self-attention mechanisms to capture the long-range relations. While existing video inpainting methods usually exploit the spatial and temporal connections in a video, our method employs a spatiotemporal transformer framework to detect the spatial connections between patches and temporal dependency between frames. As the inpainted videos usually lack high frequency details, our proposed FAST synchronously exploits the frequency domain information with a specifically designed decoder. Extensive experimental results demonstrate that our approach achieves very competitive performance and generalizes well.

We propose a novel and effective method for high-quality instance segmentation. Top-performing "detect-then-segment" approaches (e.g., Mask R-CNN) rely on region-of-interest (ROI) cropping operations to obtain the final masks, but their performance is restricted by blurry boundary and average loss weight. Here, we develop a unique perspective of image segmentation as an image matting problem. Our method, termed MMask R-CNN, enjoys two advantages: 1) Present a novel matting enhanced mask head to generate trimap-based mat-ting features as auxiliary priors, which enhance accurate estimation of alpha values for boundary pixels. 2) Explicitly represents the classification uncertainty with confidence indicator alpha matte, and design a Uncertainty-Aware Binary Cross-Entropy Loss to assign larger weights to pixels with higher uncertainty. We evaluate the proposed method through extensive experiments on the COCO dataset. The experimental results show that our method outperforms well-tuned Mask R-CNN baseline by 2.3% AP.

In multi-goal reinforcement learning (RL), efficient learning from sparse rewards remains a major challenge. One of the most successful solutions to the challenge is Hindsight Experience Replay (HER), a model-free algorithm that relabels desired goals with achieved goals. However, HER and its previous variants are still limited in efficiency and require millions of samples for training. In this paper, leveraging the power of a learned dynamics model, we propose Hindsight Experience Replay with Model-based Prediction (HERO) to further improve the sample efficiency of HER. The core technique of HERO is a two-stage value estimation algorithm combining hindsight relabeled rewards and model-based predictive rewards. To effectively model complex dynamics of robot manipulation tasks, we introduce the goal-enhanced predictive model (GPM) and the achieved-goal variance prioritization (AVP). GPM pays more attention to predicting the achieved goal in the next state, while AVP prioritizes trajectories based on the variance of achieved goals in each trajectory. In our experiments, we evaluate HERO on a set of challenging robot manipulation tasks. Empirical results demonstrate that HERO achieves significantly higher sample efficiency than previous multi-goal RL algorithms.

Boundary-based instance segmentation has drawn much attention since of its attractive efficiency. However, existing methods suffer from the difficulty in long-distance regression. In this paper, we propose a coarse-to-fine module to address the problem. Approximate boundary points are generated at the coarse stage and then features of these points are sampled and fed to a refined regressor for fine prediction. It is end-to-end trainable since differential sampling operation is well supported in the module. Furthermore, we design a holistic boundary-aware branch and introduce instance-agnostic supervision to assist regression. Equipped with ResNet-101, our approach achieves 31.7% mask AP on COCO dataset with single-scale training and testing, outperforming the baseline 1.3% mask AP with less than 1% additional parameters and GFLOPs. Experiments also show that our proposed method achieves competitive performance compared to existing boundary-based methods with a lightweight design and a simple pipeline.

We propose a novel implicit feature refinement module for high-quality instance segmentation. Existing image/video instance segmentation methods rely on explicitly stacked convolutions to refine instance features before the final prediction. In this paper, we first give an empirical comparison of different refinement strategies, which reveals that the widely-used four consecutive convolutions are not necessary. As an alternative, weight-sharing convolution blocks provides competitive performance. When such block is iterated for infinite times, the block output will eventually converge to an equilibrium state. Based on this observation, the implicit feature refinement (IFR) is developed by constructing an implicit function. The equilibrium state of instance features can be obtained by fixed-point iteration via a simulated infinite-depth network. Our IFR enjoys several advantages: 1) simulates an infinite-depth refinement network while only requiring parameters of single residual block; 2) produces high-level equilibrium instance features of global receptive field; 3) serves as a plug-and-play general module easily extended to most object recognition frameworks. Experiments on the COCO and YouTube-VIS benchmarks show that our IFR achieves improved performance on state-of-the-art image/video instance segmentation frameworks, while reducing the parameter burden (e.g. 1% AP improvement on Mask R-CNN with only 30.0% parameters in mask head). Code will be made available at \hrefhttps://github.com/lufanma/IFR.git https://github.com/lufanma/IFR.git .

Video captioning is a challenging task since it requires generating sentences describing various diverse and complex videos. Existing video captioning models lack adequate visual representation due to the neglect of the existence of gaps between videos and texts. To bridge this gap, in this paper, we propose a CLIP4Caption framework that improves video captioning based on a CLIP-enhanced video-text matching network (VTM). This framework is taking full advantage of the information from both vision and language and enforcing the model to learn strongly text-correlated video features for text generation. Besides, unlike most existing models using LSTM or GRU as the sentence decoder, we adopt a Transformer structured decoder network to effectively learn the long-range visual and language dependency. Additionally, we introduce a novel ensemble strategy for captioning tasks. Experimental results demonstrate the effectiveness of our method on two datasets: 1) on MSR-VTT dataset, our method achieved a new state-of-the-art result with a significant gain of up to 10% in CIDEr; 2) on the private test data, our method ranking 2nd place in the ACM MM multimedia grand challenge 2021: Pre-training for Video Understanding Challenge. It is noted that our model is only trained on the MSR-VTT dataset.

Existing Web service eco-systems are typically managed in a centralized manner, which hinders their further development due to inherent disadvantages such as trust issues, interest disputes, value separation and so on. The recently emerged blockchains provide distributed ledgers that enable parties who do not fully trust each other to maintain a set of global states, which provide a natural solution. Based on the INKchain, which is an open-source permissioned blockchain mechanism extending the Hyperledger Fabric, this paper proposes Trusted Decentralized Service Eco-System (T-DSES). T-DSES achieves not only fundamental functionalities of conventional systems, but also offers mechanisms to stimulate participants to bring trustworthiness to the whole system. The trustworthiness of T-DSES is realized by three strategies: reliable information of services and mashups, reliable records of participants’ rights, and reliable measurement of participants’ contributions. A customized token “SToken” is created to act as the media of value circulation. In this paper, the overall framework and detailed design of T-DSES are presented, especially including how to utilize Kubernetes to establish a cloud-based environment. A tailored Web front-end ensures the usability of operations. Over real-world data from ProgrammableWeb.com, analyses and experiments have been conducted to verify the feasibility and effectiveness of the presented approach.

his paper contributes a novel realtime multi-person motion capture algorithm using multiview video inputs. Due to the heavy occlusions and closely interacting motions in each view, joint optimization on the multiview images and multiple temporal frames is indispensable, which brings up the essential challenge of realtime efficiency. To this end, for the first time, we unify per-view parsing, cross-view matching, and temporal tracking into a single optimization framework, i.e., a 4D association graph that each dimension (image space, viewpoint and time) can be treated equally and simultaneously. To solve the 4D association graph efficiently, we further contribute the idea of 4D limb bundle parsing based on heuristic searching, followed with limb bundle assembling by proposing a bundle Kruskal's algorithm. Our method enables a realtime motion capture system running at 30fps using 5 cameras on a 5-person scene. Benefiting from the unified parsing, matching and tracking constraints, our method is robust to noisy detection due to severe occlusions and close interacting motions, and achieves high-quality online pose reconstruction quality. The proposed method outperforms state-of-the-art methods quantitatively without using high-level appearance information.

Confocal microscopy is the de-facto standard technique in bio-imaging for acquiring 3D images in the presence of tissue scattering. However, the point-scanning mechanism inherent in confocal microscopy implies that the capture speed is much too slow for imaging dynamic objects at sufficient spatial resolution and signal to noise ratio(SNR). In this paper, we propose an algorithm for super-resolution confocal microscopy that allows us to capture high-resolution, high SNR confocal images at an order of magnitude faster acquisition speed. The proposed Back-Projection Generative Adversarial Network (BPGAN) consists of a feature extraction step followed by a back-projection feedback module (BPFM) and an associated reconstruction network, these together allow for super-resolution of low-resolution confocal scans. We validate our method using real confocal captures of multiple biological specimens and the results demonstrate that our proposed BPGAN is able to achieve similar quality to high-resolution confocal scans while the imaging speed can be up to 64 times faster.

Most existing blind image quality assessment (BIQA) methods belong to supervised methods, which always need a large number of image samples and expensive subjective scores for training a quality prediction model. In this paper, we focus our attention on the unsupervised BIQA methods and put forward a novel unsupervised approach. The main idea of our method is to quantify the image quality degradation through measuring the structure, naturalness, and the perception quality variations of the distorted image from the pristine natural images. In specific, the structure variation is captured by the deviations of the image phase congruency and gradients distributions. The naturalness variation is characterized through the distributions variations of the locally mean subtracted and contrast normalized (MSCN) coefficients and the products of pairs of the adjacent MSCN coefficients. Compared with existing unsupervised methods, we initiatively introduce the perception quality measurement into the construction of unsupervised BIQA method, which is conducted by characterizing the prediction discrepancy between the image and its brain prediction based on the free-energy principle in the newly revealed brain theory. After feature extraction, we learn a pristine multivariate Gaussian (MVG) model with the extracted features from a set of pristine natural images. The quality of a new image is finally defined as the distance between its MVG model and the learned pristine MVG model. The extensive experiments conducted on LIVE, TID2013, CSIQ, Toyama, CID2013, and the Waterloo Exploration databases demonstrate that the proposed method achieves comparative prediction performance with the state-of-the-art BIQA methods.

Most existing unsupervised video hashing methods generate binary codes by using RNNs in a deterministic manner, which fails to capture the dominant latent variation of videos. In addition, RNN-based video hashing methods suffer the content forgetting of early input frames due to the sequential processing inherency of RNNs, which is detrimental to global information capturing. In this work, we propose an unsupervised variational video hashing (UVVH) method for scalable video retrieval. Our UVVH method aims to capture the salient and global information in a video. Specifically, we introduce a variational autoencoder to learn a probabilistic latent representation of the salient factors of video variations. To better exploit the global information of videos, we design a 1D-CNN-LSTM model. The 1D-CNN-LSTM model processes long frame sequences in a parallel and hierarchical way, and exploits the correlations between frames to reconstruct the frame-level features. As a consequence, the learned hash functions can produce reliable binary codes for video retrieval. We conduct extensive experiments on three widely used benchmark datasets, FCVID, ActivityNet and YFCC to validate the effectiveness of our proposed approach.

Opinion-unaware blind image quality assessment (OU BIQA) refers to establishing a blind quality prediction model without using the expensive subjective quality scores, which is a highly promising direction in the BIQA research. In this article, we focus on OU BIQA and propose a novel OU BIQA method. Specifically, in our proposed method, we deeply investigate the natural scene statistics (NSS) and the perceptual characteristics of the human brain for visual perception. Accordingly, a set of quality-aware NSS and perceptual characteristics-related features are designed to characterize the image quality effectively. For inferring the image quality, we learn a pristine multivariate Gaussian (MVG) model on a collection of pristine images, which serves as the reference information for quality evaluation. At last, the quality of a new given image is defined by measuring the divergence between its MVG model and the learned pristine MVG model. Thorough experiments performed on seven popular image databases demonstrate that the proposed OU BIQA method delivers superior performance to the state-of-the-art OU BIQA methods. The Matlab source code of the proposed method will be made publicly available at https://github.com/YT2015?tab=;repositories.

In this paper, we propose a Neighborhood Preserving Hashing (NPH) method for scalable video retrieval in an unsupervised manner. Unlike most existing deep video hashing methods which indiscriminately compress an entire video into a binary code, we embed the spatial-temporal neighborhood information into the encoding network such that the neighborhood-relevant visual content of a video can be preferentially encoded into a binary code under the guidance of the neighborhood information. Specifically, we propose a neighborhood attention mechanism which focuses on partial useful content of each input frame conditioned on the neighborhood information. We then integrate the neighborhood attention mechanism into an RNN-based reconstruction scheme to encourage the binary codes to capture the spatial-temporal structure in a video which is consistent with that in the neighborhood. As a consequence, the learned hashing functions can map similar videos to similar binary codes. Extensive experiments on three widely-used benchmark datasets validate the effectiveness of our proposed approach.

In the past a few years, we witnessed rapid advancement in face super-resolution from very low resolution(VLR) images. However, most of the previous studies focus on solving such problem without explicitly considering the impact of severe real-life image degradation (e.g. blur and noise). We can show that robustly recover details from VLR images is a task beyond the ability of current state-of-the-art method. In this paper, we borrow ideas from "facial composite" and propose an alternative approach to tackle this problem. We endow the degraded VLR images with additional cues by integrating existing face components from multiple reference images into a novel learning pipeline with both low level and high level semantic loss function as well as a specialized adversarial based training scheme. We show that our method is able to effectively and robustly restore relevant facial details from 16x16 images with extreme degradation. We also tested our approach against real-life images and our method performs favorably against previous methods.

Service ecosystems, consisting of various kinds of services and mashups, usually keep evolving over time.Existing works on the evolution of service ecosystems focus on either evaluating the impacts of single services' changes on the usage of services and the stability of the whole ecosystem, or discovering co-occurrence relationship between services, but fail to disclose any knowledge from the aspect of the evolution of service composition patterns.Based on our previous work, this paper moves one step further, revealing the latent service composition trends in a service ecosystem and providing more distinct explanation of different topic evolution patterns.A novel methodology, named Extended Dependency-Compensated Service Co-occurrence LDA (EDC-SeCo-LDA), is developed to calculate the directed dependencies between different topics and build topic evolution graph.The evolution trend

This paper investigates the vulnerability of the deep prior used in deep learning based image restoration. In particular, the image super-resolution, which relies on the strong prior information to regularize the solution space and plays important roles in the image pre-processing for future viewing and analysis, is shown to be vulnerable to the well-designed adversarial examples. We formulate the adversarial example generation process as an optimization problem, and given super-resolution model three different types of attack are designed based on the subsequent tasks: (i) style transfer attack; (ii) classification attack; (iii) caption attack. Another interesting property of our design is that the attack is hidden behind the super-resolution process, such that the utilization of low resolution images is not significantly influenced. We show that the vulnerability to adversarial examples could bring risks to the pre-processing modules such as super-resolution deep neural network, which is also of paramount significance for the security of the whole system. Our results also shed light on the potential security issues of the pre-processing modules, and raise concerns regarding the corresponding countermeasures for adversarial examples.

Convolutional neural network (CNN) based face detectors are inefficient in handling faces of diverse scales. They rely on either fitting a large single model to faces across a large scale range or multi-scale testing. Both are computationally expensive. We propose Scale-aware Face Detection (SAFD) to handle scale explicitly using CNN, and achieve better performance with less computation cost. Prior to detection, an efficient CNN predicts the scale distribution histogram of the faces. Then the scale histogram guides the zoom-in and zoom-out of the image. Since the faces will be approximately in uniform scale after zoom, they can be detected accurately even with much smaller CNN. Actually, more than 99% of the faces in AFW can be covered with less than two zooms per image. Extensive experiments on FDDB, MALF and AFW show advantages of SAFD.

In this paper, we propose a novel method for automatic depth estimation from color images using parameter transfer. By modeling the correlation between color images and their depth maps with a set of parameters, we get a database of parameter sets. Given an input image, we extract the high-level features to find the best matched image sets from the database. Then the set of parameters corresponding to the best match are used to estimate the depth of the input image. Compared with the past learning-based methods, our trained model consists only of trained features and parameter sets, which occupy little space. We evaluate our depth estimation method on several benchmark RGB-D (RGB + depth) data sets. The experimental results are comparable to the state-of-the-art results, while the model size is very small and very suitable for mobile devices, demonstrating the promising performance of our proposed method.

Cascade has been widely used in face detection, where classifier with low computation cost can be firstly used to shrink most of the background while keeping the recall. The cascade in detection is popularized by seminal Viola-Jones framework and then widely used in other pipelines, such as DPM and CNN. However, to our best knowledge, most of the previous detection methods use cascade in a greedy manner, where previous stages in cascade are fixed when training a new stage. So optimizations of different CNNs are isolated. In this paper, we propose joint training to achieve end-to-end optimization for CNN cascade. We show that the back propagation algorithm used in training CNN can be naturally used in training CNN cascade. We present how jointly training can be conducted on naive CNN cascade and more sophisticated region proposal network (RPN) and fast R-CNN. Experiments on face detection benchmarks verify the advantages of the joint training.

Clustering data streams has become a hot topic in the era of big data. Driven by the ever increasing volume, velocity and variety of data, more efficient algorithms for clustering large-scale complex data streams are needed. In this paper, we present a parallel algorithm called PaStream, which is based on advanced Graphics Processing Unit (GPU) and follows the online-offline framework of CluStream. Our approach can achieve hundreds of times speedup on high-speed and high-dimensional data streams compared with CluStream. It can also discover clusters with arbitrary shapes and handle outliers properly. The efficiency and scalability of PaStream are demonstrated through comprehensive experiments on synthetic and standard benchmark datasets with various problem factors.

In recent years, water quality prediction has attracted many attentions of governments and researchers. The safety of water quality seriously affects the human health, fishery economy and agricultural activities. If an early prediction to the water quality with an acceptable accuracy can be achieved, the negative impacts will be minimized or even be avoided. Many researchers have applied artificial neural networks (ANNs) to build the water quality models for there is a complicated non-linear relation between the prediction variables and measured inputs. However the ANN models are easy to be over-fitting for training them needs a large of samples. As the objective of this study, artificial neural network and Markov chain approach are used to develop a new hybrid methodology for predicting the biochemical oxygen demand which is the main indicator of water quality. ANN produces the primary values and then the results are modified by three regression methods using the Markov transitional probability matrices respectively. We use a 27-year water quality data set of Tolo Harbor which only has a total of 439 samples to test our method. The results are validated and a better prediction accuracy of the new ANN-Markov Chain Methodology is demonstrated through three criteria.

The Internet significantly reshapes traditional commerce mode, making online business ubiquitous and indispensable. Others' opinions are influential when we make a decision, which is complicated by tremendous and misleading information. According to opinion mining techniques, we propose a novel conceptual framework which is designed to exploit the potentials of opinion mining in e-commerce domain. Based on the various text corpora we crawled from targeted internet source, we utilize the sentiment analysis principle and algorithm to provide convenient and assistant services, such as opinion spam detection, opinion search and retrieval, opinion summary, opinion question answering and opinion recommendation, to e-commerce participants such as potential customers, manufacturers, third-party traders or retailers, and regulators. Besides, the system introduces a novel NLP toolkits excelling in handling the Chinese corpora than most of previous alike work. The system is especially helpful to regulators by providing useful applications which can be utilized to make administrating decisions.

The problem of China's coastal water pollution is becoming more and more serious. Thus the management of sewage outfall with technology methods is gaining more attention. WebGIS technology such as ArcGIS Server, ArcGIS Online and Flex, etc. may be of great help. We can establish a B/S structure system with the advantages of easy building up and function expansion, high information sharing and strong user interaction capability, low cost, and so on. The approach to design and Implement a Sewage Outfall Management System based on WebGIS technology is discussed in this article. The results show that this system can better manage the outfall information and provide support to decision-making than traditional ones, and it also has broad application prospects in other fields of environmental protection. This system has been applied to our project so far and just went through the interim review.

Support vector machine (SVM) and its derivative algorithms have been increasingly used to predict algal blooms recently. However, its computation complexity remains an annoying problem. To improve the time cost of SVM, a hybrid approach is proposed in this paper based on Partial Least Square (PLS) feature extraction and Core Vector Machine Regression (CVR) algorithm. We describe the principle of our algorithm and the implementation steps in detail. Based on the biweekly data gathered from Tolo Harbour, Hong Kong, some comparative analysis of the performance of PLS-CVR and other algorithms are presented. We develop these prediction models with different lead time (7-day and 14-day) to study further. The results indicate that the use of biweekly data can simulate the general trend of algal biomass reasonably. The experimental results show that our algorithm can reduce the time cost significantly compared to conventional SVM algorithm while maintaining a satisfactory accuracy.

Machine learning (ML) techniques such as artificial neural network (ANN) and support vector machine (SVM) have been increasingly used to predict harmful algal blooms (HABs). In this paper, we use the biweekly data in Tolo Harbour, Hong Kong, and choose several machine learning methods to develop prediction models of algal blooms. Three different kinds of models are designed based on back-propagation (BP) neural network, generalized regression neural network (GRNN) and support vector machine (SVM) respectively. The experimental results show that the improved BP algorithm and SVM work better than GRNN methods, and the models based on SVM present the best performance in terms of goodness-of-fit measures, but need to be further improved in the running time. We develop these prediction models with different lead time (7-day and 14-day) to study further. The results indicate that the use of biweekly data can simulate the general trend of algal biomass reasonably, but it is not ideally suited for exact predictions. The use of higher frequency data may improve the accuracy of the predictions.

This paper presents a comprehensive review of Cyberinfrastructure (CI), an emerging collaborative research environment, including its representative applications in four science communities around the world. An in-depth analysis is also conducted to reveal the key functions and desired features that can be expected from modern CI systems.

Nowadays, rapid evolution of computers and mobile devices has caused the explosive increase in network traffic. So it becomes more and more necessary to archive network traffic for analyzing network events and a lot of emerging applications. Compression is fundamental for traffic archival solution to save the storage space, and indexing is effective to accelerate search queries for archive of traffic data. In this paper, we propose BreadZip (blocks row-reordering and adaptive index zip), a combination of initial traffic data and index compression. BreadZip has three main advantages. 1) to improve compressing efficiency and reduce memory footprint, traffic data is reordered in sequence and divided into fixed-size blocks; 2) to accelerate queries, an improved bitmap indexes with smaller volume than traditional will be introduced; 3) to save space, both traffic blocks and bitmap indexes are compressed in different simple run-length encoding methods respectively. Finally, our empirical results on network traffic from CAIDA (Cooperative Association for Internet Data Analysis) show that our solution can significantly reduce the volume of traffic data, while simultaneously preserving the ability to perform selectively queries with response times in seconds.

Based on the introduction of the concept of the 4th party logistics and the decision supporting system in its operation, the directed graph model with multi dimension weight was established for its optimization problem of how to select the route, the transportation method and the third party logistics provider concurrently. Genetic Algorithm is used to solve this directed graph model. An adaptive length-variant coding method is employed to represent the individual. The initial group is generated at random. After crossover and mutation are done, individuals are selected according to the rank they obtain when comparison is taken in the sense of fitness, of course randomly and elitism is applied. The results of experiments show that this Genetic algorithm can help to get the optimal solution to the directed graph model and solve the optimization problem in the operation of 4th party logistics.

With more and more manufacturers and tradesmen put logistic outsourcing as a viable option for their business, fourth party logistics (4PL) has become a hot topic currently. The concept and the evaluation of 4PL are introduced. Focusing on the e-business environment, the character, the operation mode and the support relation are analyzed. Then, based on a logistic enterprise, structure and functionality of 4PL with reference to the TY-4PL software platform are discussed, critical success factors for realization project are named.

With the Increased competition in the transport market, the new international logistics service provider must develop solutions tailored to meet the unique and special needs of each customer. Building upon the foundation of 3rd party outsourcing, 4PL emerges and acts as an electronic intermediator that connects suppliers, producers, carriers and customers to dynamic supply chain networks. Of such transportation networks the operative planning task of the hub-and-spoke network is a challenging task for the management. In particular, the transport management has to decide where all quantifies within the transportation network flow over the hub from or to the depots, or whether a hybrid hub-and-spoke network is preferred in which direct transports also take place. Upon these, Optimization model is studied and a 4PL optimization system for these operative planning tasks is developed and applied to a real case of a Chinese logistic service provider. The result of the case indicates that the model and algorithm presented in the paper can provide a reasonable supply chain scheduling scheme.

Business process re-engineering (BPR) is the philosophy, which re-designs the existing processes according to the requirement of internal processes and external customers. But there was not an efficient model for enterprise reconstructions. Quality function deployment (QFD) offers strengths complementary to BPR and it provides a new way to analyze BPR. Traditionally, QFD approach has been used as a tool for defining new products, as well as for improving existing products. In substance, QFD builds the relationship between "target" (what to do) and "measure" (how to do). In this paper, we introduce QFD methodology, are presented to perform BPR.

Quality function development (QFD) is an important methodology for concurrent engineering. QFD contacts the customer requirements and production process. In distributed environment, customer requirements, multiple team members and process are geographically, culturally, and functionally diverse. It requires cooperation and conflict resolution among design, production and sales process. In this paper, we introduce a multi-agent-based distributed QFD framework, which orients concurrent engineering (MAS-CEQFD). In this framework, QFD core module, manufacture module, distributed TM module and market module are presented. Agents are used to emulate the entities, i.e. various module and their internal departments. The framework facilitates the development of the product design and the production process.

Based on the analysis of various ways of solving job-shop scheduling problems, the mathematical model of job-shop problem is introduced. A kind of genetic algorithm is used to solve it in this paper. Firstly, the chromosomes are encoded, and the population size is premised for the optimized goal of job-shop problem, which are the keys of the method. Secondly, the fitness function is designed. After using selection, crossover and mutation operator, and then elitist strategy to prevent the premature convergence, a best or satisfactory scheduling path can be found. The executing results using the presented algorithms in certain project case are shown in the paper.