Ensuring alignment with human values is essential for modern large language models (LLMs), especially amid growing concerns around AI safety and social impact. Yet achieving such alignment remains challenging due to the limited, noisy, and often conflicting nature of human feedback from diverse annotators. Most existing approaches, such as Direct Preference Optimization (DPO), assume consistent and conflict-free supervision, overlooking the ambiguity, inconsistency, and value trade-offs inherent in real-world preferences—often leading to reduced robustness and exclusion of minority views. To address this, we propose FGD-Align, a novel pluralistic alignment framework grounded in Fuzzy Group Decision-Making theory. Our approach rigorously models and aggregates human preferences while retaining the complexity of real-world value trade-offs. Unlike traditional methods that rely on coarse-grained preference pairs, FGD-Align introduces fuzzy preference modeling via triangular fuzzy numbers to capture nuanced, multi-criteria human judgments. We further develop a new training objective, Probabilistic Fuzzy DPO, which incorporates fuzzy preference strength as adaptive loss weights and gradient filters, enhancing robustness to ambiguity and inconsistency in feedback. Comprehensive experiments demonstrate that FGD-Align consistently outperforms both DPO variants and advanced preference aggregation methods in terms of preference accuracy and robustness to ambiguity. It achieves superior alignment stability and better preserves minority preferences, all with minimal computational overhead. Our work bridges the gap between algorithmic tractability and the nuanced landscape of human values, enabling more scalable, inclusive, and socially-aware AI alignment.

Accurate trajectory prediction has prominent significance in autonomous driving scenarios. Most existing methods predict the trajectory of an agent by learning its interaction with other agents and the map within the scenario. However, the heterogeneous distribution of these elements across different geographical scenarios is always ignored. Thus, trajectory predictors might struggle to generalize well when deployed in different geographical scenarios. To bridge the cross-geography gap, in this paper, we propose a plug-and-play self-training pipeline, termed STraj, for cross-geography trajectory prediction. STraj comprises three progressive steps: pseudo label (i.e., time-series trajectory) generation, update, and utilization. First, to generate pseudo labels that generalize to the cross-geography scenarios, STraj pre-trains the predictor through the complementary agent and map augmentations. Second, to facilitate the stable training of the predictor, we design a specific pseudo label update strategy. This strategy selects high-consistency pseudo trajectories from the current and historical epochs to supervise the target domain samples. Third, with generated pseudo trajectories, we introduce trajectory-induced contrastive learning to mitigate the representation bias of cross-geography agents. Extensive experiment results on various cross-geography trajectory prediction benchmarks demonstrate the effectiveness of STraj.

Chenxi Huang, Shaotian Yan, Liang Xie, Binbin Lin, Sinan Fan, Yue Xin, Deng Cai, Chen Shen, Jieping Ye. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Graph Convolutional Networks (GCNs) are powerful tools for learning representations in graph-structured data. However, their performance tends to degrade with increased model depth due to over-smoothing. Although previous studies attribute degradation to over-smoothing, this work identifies the mutually reinforcing effects of over-smoothing and gradient vanishing as the root cause. In this paper, we propose SkipNode, a plug-and-play module that mitigates degradation in deep GCNs. SkipNode introduces node-sampling in each convolutional layer to selectively skip convolutions, preventing over-smoothing by reducing the depth experienced by specific nodes and facilitating gradient backpropagation. We demonstrate both theoretically and experimentally that SkipNode effectively curtails over-smoothing and gradient vanishing, improving deep GCN performance across diverse tasks. Extensive evaluations show SkipNode's robustness and superior performance over state-of-the-art (SOTA) baselines, establishing it as a practical solution for training deep GCNs.

Video depth estimation has long been hindered by the scarcity of consistent and scalable ground truth data, leading to inconsistent and unreliable results. In this paper, we introduce Depth Any Video, a model that tackles the challenge through two key innovations. First, we develop a scalable synthetic data pipeline, capturing real-time video depth data from diverse virtual environments, yielding 40,000 video clips of 5-second duration, each with precise depth annotations. Second, we leverage the powerful priors of generative video diffusion models to handle real-world videos effectively, integrating advanced techniques such as rotary position encoding and flow matching to further enhance flexibility and efficiency. Unlike previous models, which are limited to fixed-length video sequences, our approach introduces a novel mixed-duration training strategy that handles videos of varying lengths and performs robustly across different frame rates—even on single frames. At inference, we propose a depth interpolation method that enables our model to infer high-resolution video depth across sequences of up to 150 frames. Our model outperforms all previous generative depth models in terms of spatial accuracy and temporal consistency. The code and model weights are open-sourced.

Local geometry-controllable computer-aided design (CAD) generation aims to modify local parts of CAD models automatically, enhancing design efficiency. It also ensures that the shapes of newly generated local parts follow user-specific geometric instructions (e.g., an isosceles right triangle or a rectangle with one corner cut off). However, existing methods encounter challenges in achieving this goal. Specifically, they either lack the ability to follow textual instructions or are unable to focus on the local parts. To address this limitation, we introduce GeoCAD, a user-friendly and local geometry-controllable CAD generation method. Specifically, we first propose a complementary captioning strategy to generate geometric instructions for local parts. This strategy involves vertex-based and VLLM-based captioning for systematically annotating simple and complex parts, respectively. In this way, we caption $\sim$221k different local parts in total. In the training stage, given a CAD model, we randomly mask a local part. Then, using its geometric instruction and the remaining parts as input, we prompt large language models (LLMs) to predict the masked part. During inference, users can specify any local part for modification while adhering to a variety of predefined geometric instructions. Extensive experiments demonstrate the effectiveness of GeoCAD in generation quality, validity and text-to-CAD consistency.

Emerging reasoning LLMs such as OpenAI-o1 and DeepSeek-R1 have achieved strong performance on complex reasoning tasks by generating long chain-of-thought (CoT) traces. However, these long CoTs result in increased token usage, leading to higher inference latency and memory consumption. As a result, balancing accuracy and reasoning efficiency has become essential for deploying reasoning LLMs in practical applications. Existing long-to-short (Long2Short) methods aim to reduce inference length but often sacrifice accuracy, revealing a need for an approach that maintains performance while lowering token costs. To address this efficiency-accuracy tradeoff, we propose TokenSqueeze, a novel Long2Short method that condenses reasoning paths while preserving performance and relying exclusively on self-generated data. First, to prevent performance degradation caused by excessive compression of reasoning depth, we propose to select self-generated samples whose reasoning depth is adaptively matched to the complexity of the problem. To further optimize the linguistic expression without altering the underlying reasoning paths, we introduce a distribution-aligned linguistic refinement method that enhances the clarity and conciseness of the reasoning path while preserving its logical integrity. Comprehensive experimental results demonstrated the effectiveness of TokenSqueeze in reducing token usage while maintaining accuracy. Notably, DeepSeek‑R1‑Distill‑Qwen‑7B fine-tuned by using our proposed method achieved a 50\% average token reduction while preserving accuracy on the MATH500 benchmark. TokenSqueeze exclusively utilizes the model's self-generated data, enabling efficient and high-fidelity reasoning without relying on manually curated short-answer datasets across diverse applications. Our code is available at \url{https://github.com/zhangyx1122/TokenSqueeze}.

Lane is critical in the vision navigation system of intelligent vehicles. Naturally, the lane is a traffic sign with high-level semantics, whereas it owns the specific local pattern which needs detailed low-level features to localize accurately. Using different feature levels is of great importance for accurate lane detection, but it is still under-explored. On the other hand, current lane detection methods still struggle to detect complex dense lanes, such as Y-shape or fork-shape. In this work, we present Cross Layer Refinement Network aiming at fully utilizing both high-level and low-level features in lane detection. In particular, it first detects lanes with high-level semantic features and then performs refinement based on low-level features. In this way, we can exploit more contextual information to detect lanes while leveraging local-detailed features to improve localization accuracy. We present Fast-ROIGather to gather global context, which further enhances the representation of lane features. To detect dense lanes accurately, we propose Correlation Discrimination Module (CDM) to discriminate the correlation of dense lanes, enabling nearly cost-free high-quality dense lane prediction. In addition to our novel network design, we introduce LineIoU loss which regresses lanes as a whole unit to improve localization accuracy. Experiments demonstrate our approach significantly outperforms the state-of-the-art lane detection methods.

NeRF-Det has achieved impressive performance in indoor multi-view 3D detection by innovatively utilizing NeRF to enhance representation learning. Despite its notable performance, we uncover three decisive shortcomings in its current design, including semantic ambiguity, inappropriate sampling, and insufficient utilization of depth supervision. To combat the aforementioned problems, we present three corresponding solutions: 1) Semantic Enhancement. We project the freely available 3D segmentation annotations onto the 2D plane and leverage the corresponding 2D semantic maps as the supervision signal, significantly enhancing the semantic awareness of multi-view detectors. 2) Perspective-Aware Sampling. Instead of employing the uniform sampling strategy, we put forward the perspective-aware sampling policy that samples densely near the camera while sparsely in the distance, more effectively collecting the valuable geometric clues. 3) Ordinal Residual Depth Supervision. As opposed to directly regressing the depth values that are difficult to optimize, we divide the depth range of each scene into a fixed number of ordinal bins and reformulate the depth prediction as the combination of the classification of depth bins as well as the regression of the residual depth values, thereby benefiting the depth learning process. The resulting algorithm, NeRF-Det++, has exhibited appealing performance in the ScanNetV2 and ARKITScenes datasets. Notably, in ScanNetV2, NeRF-Det++ outperforms the competitive NeRF-Det by +1.9% in mAP $\text{@}0.25$ and +3.5% in mAP $\text{@}0.50$ . The code will be publicly available at https://github.com/mrsempress/NeRF-Detplusplus.

The storage location assignment problem for export containers (EC-SLAP) is crucial to the efficiency of cargo turnover in ports. Existing methods fall short in real-world applications due to the challenges of the unpredictability of container arrival sequences and the large-scale problem. We propose a new buffer-wise framework based on hierarchical reinforcement learning for EC-SLAP, aimed at optimizing container turnover efficiency. The framework comprises two processes: 1)Ranking Agentranks containers in the buffer, reducing the uncertainty of random arrival sequences compared to the immediate assignment. 2)Assigning Agentsassign storage locations in a two-step process by block and slot, diminishing the dimensionality of the large-scale discrete action space. We iteratively optimize agents by asynchronously obtaining rewards from the environment. In addition, to address the challenge of sparse rewards in long-sequence decision-making, we have developed a novel immediate reward function to enhance learning efficiency and accelerate convergence. We propose a new large-scale dataset, NZP-SLAD, collected from real-world historical data from the terminal operating system of Ningbo-Zhoushan Port and develop a realistic container terminal simulator. We conducted numerous offline simulations and tests with this dataset. The experimental results demonstrate that our proposed method achieves rapid convergence and significantly surpasses expert methods used in real-world production.

Contrastive Language-Image Pre-training (CLIP) has demonstrated impressive capabilities in open-vocabulary classification. The class token in the image encoder is trained to capture the global features to distinguish different text descriptions supervised by contrastive loss, making it highly effective for single-label classification. However, it shows poor performance on multi-label datasets because the global feature tends to be dominated by the most prominent class and the contrastive nature of softmax operation aggravates it. In this study, we observe that the multi-label classification results heavily rely on discriminative local features but are overlooked by CLIP. As a result, we dissect the preservation of patch-wise spatial information in CLIP and proposed a local-to-global framework to obtain image tags. It comprises three steps: (1) patch-level classification to obtain coarse scores; (2) dual-masking attention refinement (DMAR) module to refine the coarse scores; (3) class-wise reidentification (CWR) module to remedy predictions from a global perspective. This framework is solely based on frozen CLIP and significantly enhances its multi-label classification performance on various benchmarks without dataset-specific training. Besides, to comprehensively assess the quality and practicality of generated tags, we extend their application to the downstream task, i.e., weakly supervised semantic segmentation (WSSS) with generated tags as image-level pseudo labels. Experiments demonstrate that this classify-then-segment paradigm dramatically outperforms other annotation-free segmentation methods and validates the effectiveness of generated tags. Our code is available at https://github.com/linyq2117/TagCLIP.

Semi-supervised learning aims to leverage numerous unlabeled data to improve the model performance. Current semi-supervised 3D object detection methods typically use a teacher to generate pseudo labels for a student, and the quality of the pseudo labels is essential for the final performance. In this paper, we propose PatchTeacher, which focuses on partial scene 3D object detection to provide high-quality pseudo labels for the student. Specifically, we divide a complete scene into a series of patches and feed them to our PatchTeacher sequentially. PatchTeacher leverages the low memory consumption advantage of partial scene detection to process point clouds with a high-resolution voxelization, which can minimize the information loss of quantization and extract more fine-grained features. However, it is non-trivial to train a detector on fractions of the scene. Therefore, we introduce three key techniques, i.e., Patch Normalizer, Quadrant Align, and Fovea Selection, to improve the performance of PatchTeacher. Moreover, we devise PillarMix, a strong data augmentation strategy that mixes truncated pillars from different LiDAR scans to generate diverse training samples and thus help the model learn more general representation. Extensive experiments conducted on Waymo and ONCE datasets verify the effectiveness and superiority of our method and we achieve new state-of-the-art results, surpassing existing methods by a large margin. Codes are available at https://github.com/LittlePey/PTPM.

Human body orientation estimation (HBOE) aims to estimate the orientation of a human body relative to the camera’s frontal view. Despite recent advancements in this field, there still exist limitations in achieving fine-grained results. We identify certain defects and propose corresponding approaches as follows: 1). Existing datasets suffer from non-uniform angle distributions, resulting in sparse image data for certain angles. To provide comprehensive and high-quality data, we introduce RMOS (Rendered Model Orientation Set), a rendered dataset comprising 150K accurately labeled human instances with a wide range of orientations. 2). Directly using one-hot vector as labels may overlook the similarity between angle labels, leading to poor supervision. And converting the predictions from radians to degrees enlarges the regression error. To enhance supervision, we employ Laplace smoothing to vectorize the label, which contains more information. For fine-grained predictions, we adopt weighted Smooth-L1-loss to align predictions with the smoothed-label, thus providing robust supervision. 3). Previous works ignore body-part-specific information, resulting in coarse predictions. By employing local-window self-attention, our model could utilize different body part information for more precise orientation estimations. We validate the effectiveness of our method in the benchmarks with extensive experiments and show that our method outperforms state-of-the-art. Project is available at: https://github.com/Whalesong-zrs/Towards-Fine-grained-HBOE.

Training deep models for LiDAR semantic segmentation is challenging due to the inherent sparsity of point clouds. Utilizing temporal data is a natural remedy against the spar-sity problem as it makes the input signal denser. However, previous multi-frame fusion algorithms fall short in utilizing sufficient temporal information due to the memory constraint, and they also ignore the informative temporal images. To fully exploit rich information hidden in long-term temporal point clouds and images, we present the Temporal Aggre-gation Network, termed TASeg. Specifically, we propose a Temporal LiDAR Aggregation and Distillation (TLAD) algorithm, which leverages historical priors to assign dif-ferent aggregation steps for different classes. It can largely reduce memory and time overhead while achieving higher accuracy. Besides, TLAD trains a teacher injected with gt priors to distill the model, further boosting the performance. To make full use of temporal images, we design a Temporal Image Aggregation and Fusion (TIAF) module, which can greatly expand the camera FOVand enhance the present features. Temporal LiDAR points in the camera FOV are used as mediums to transform temporal image features to the present coordinate for temporal multi-modal fusion. Moreover, we develop a Static-Moving Switch Augmentation (SMSA) algorithm, which utilizes sufficient temporal information to enable objects to switch their motion states freely, thus greatly increasing static and moving training samples. Our TASeg ranks 1st††the date of CVPR deadline, i.e., 2023-11-18 07:59 AM UTC. on three challenging tracks, i.e., SemanticKITTI single-scan track, multi-scan track and nuScenes LiDAR segmentation track, strongly demonstrating the superiority of our method. Codes are available at https://github.com/LittlePey/TASeg.

In the context of autonomous driving, the significance of effective feature learning is widely acknowledged. While conventional 3D self-supervised pretraining methods have shown widespread success, most methods follow the ideas originally designed for 2D images. In this paper, we present UniPAD, a novel self-supervised learning paradigm applying 3D volumetric differentiable rendering. UniPAD implicitly encodes 3D space, facilitating the reconstruction of continuous 3D shape structures and the intricate appear-ance characteristics of their 2D projections. The flexibil-ity of our method enables seamless integration into both 2D and 3D frameworks, enabling a more holistic compre-hension of the scenes. We manifest the feasibility and effectiveness of UniPAD by conducting extensive experiments on various 3D perception tasks. Our method significantly improves lidar-, camera-, and lidar-camera-based baseline by 9.1, 7.7, and 6.9 NDS, respectively. Notably, our pretraining pipeline achieves 73.2 NDS for 3D object detection and 79.4 mIoU for 3D semantic segmentation on the nuScenes validation set, achieving state-of-the-art results in comparison with previous methods.

Recent self-training techniques have shown notable improvements in unsupervised domain adaptation for 3D object detection (3D UDA). These techniques typically select pseudo labels, i.e., 3D boxes, to supervise models for the target domain. However, this selection process inevitably introduces unreliable 3D boxes, in which 3D points cannot be definitively assigned as foreground or background. Previous techniques mitigate this by reweighting these boxes as pseudo labels, but these boxes can still poison the training process. To resolve this problem, in this paper, we propose a novel pseudo label refinery framework. Specifically, in the selection process, to improve the reliability of pseudo boxes, we propose a complementary augmentation strategy. This strategy involves either removing all points within an unre-liable box or replacing it with a high-confidence box. More-over, the point numbers of instances in high-beam datasets are considerably higher than those in low-beam datasets, also degrading the quality of pseudo labels during the training process. We alleviate this issue by generating additional proposals and aligning RoI features across different domains. Experimental results demonstrate that our method effectively enhances the quality of pseudo labels and consistently surpasses the state-of-the-art methods on six autonomous driving benchmarks. Code will be available at https://github.com/Zhanwei-Z/PERE.

Large Language Models (LLMs) tend to prioritize adherence to user prompts over providing veracious responses, leading to the sycophancy issue. When challenged by users, LLMs tend to admit mistakes and provide inaccurate responses even if they initially provided the correct answer. Recent works propose to employ supervised fine-tuning (SFT) to mitigate the sycophancy issue, while it typically leads to the degeneration of LLMs' general capability. To address the challenge, we propose a novel supervised pinpoint tuning (SPT), where the region-of-interest modules are tuned for a given objective. Specifically, SPT first reveals and verifies a small percentage (<5%) of the basic modules, which significantly affect a particular behavior of LLMs. i.e., sycophancy. Subsequently, SPT merely fine-tunes these identified modules while freezing the rest. To verify the effectiveness of the proposed SPT, we conduct comprehensive experiments, demonstrating that SPT significantly mitigates the sycophancy issue of LLMs (even better than SFT). Moreover, SPT introduces limited or even no side effects on the general capability of LLMs. Our results shed light on how to precisely, effectively, and efficiently explain and improve the targeted ability of LLMs.

Large Language Models (LLM) based agents have shown promise in autonomously completing tasks across various domains, e.g., robotics, games, and web navigation. However, these agents typically require elaborate design and expert prompts to solve tasks in specific domains, which limits their adaptability. We introduce AutoManual, a framework enabling LLM agents to autonomously build their understanding through interaction and adapt to new environments. AutoManual categorizes environmental knowledge into diverse rules and optimizes them in an online fashion by two agents: 1) The Planner codes actionable plans based on current rules for interacting with the environment. 2) The Builder updates the rules through a well-structured rule system that facilitates online rule management and essential detail retention. To mitigate hallucinations in managing rules, we introduce a *case-conditioned prompting* strategy for the Builder. Finally, the Formulator agent compiles these rules into a comprehensive manual. The self-generated manual can not only improve the adaptability but also guide the planning of smaller LLMs while being human-readable. Given only one simple demonstration, AutoManual significantly improves task success rates, achieving 97.4\% with GPT-4-turbo and 86.2\% with GPT-3.5-turbo on ALFWorld benchmark tasks. The code is available at https://github.com/minghchen/automanual.

Federated Learning (FL) is a prevalent machine learning paradigm designed to address challenges posed by heterogeneous client data while preserving data privacy. Unlike distributed training, it typically orchestrates resource-constrained edge devices to communicate via a low-bandwidth communication network with a central server. This urges the development of more computation and communication efficient training algorithms. In this paper, we propose an efficient FL paradigm, where the local models in the clients are trained with low-precision operations and communicated with the server in low precision format, while only the model aggregation in the server is performed with high-precision computation. We surprisingly find that high precision models can be recovered from the low precision local models with proper aggregation in the server. In this way, both the workload in the client-side and the communication cost can be significantly reduced. We theoretically show that our proposed paradigm can converge to the optimal solution as the training goes on, which demonstrates that low precision local training is enough for FL. Our paradigm can be integrated with existing FL algorithms flexibly. Experiments across extensive benchmarks are conducted to showcase the effectiveness of our proposed method. Notably, the models trained by our method with the precision as low as 8 bits are comparable to those from the full precision training. As a by-product, we show that low precision local training can relieve the over-fitting issue in local training, which under heterogeneous client data can cause the client models drift further away from each other and lead to the failure in model aggregation. Code is released at https://github.com/digbangbang/LPT-FL.

While large language models (LLMs) showcase unprecedented capabilities, they also exhibit certain inherent limitations when facing seemingly trivial tasks. A prime example is the recently debated "reversal curse", which surfaces when models, having been trained on the fact "A is B", struggle to generalize this knowledge to infer that "B is A". In this paper, we examine the manifestation of the reversal curse across various tasks and delve into both the generalization abilities and the problem-solving mechanisms of LLMs. This investigation leads to a series of significant insights: (1) LLMs are able to generalize to "B is A" when both A and B are presented in the context as in the case of a multiple-choice question. (2) This generalization ability is highly correlated to the structure of the fact "A is B" in the training documents. For example, this generalization only applies to biographies structured in "[Name] is [Description]" but not to "[Description] is [Name]". (3) We propose and verify the hypothesis that LLMs possess an inherent bias in fact recalling during knowledge application, which explains and underscores the importance of the document structure to successful learning. (4) The negative impact of this bias on the downstream performance of LLMs can hardly be mitigated through training alone. Based on these intriguing findings, our work not only presents a novel perspective for interpreting LLMs' generalization abilities from their intrinsic working mechanism but also provides new insights for the development of more effective learning methods for LLMs.

As the development and application of Large Language Models (LLMs) continue to advance rapidly, enhancing their trustworthiness and aligning them with human preferences has become a critical area of research. Traditional methods rely heavily on extensive data for Reinforcement Learning from Human Feedback (RLHF), but representation engineering offers a new, training-free approach. This technique leverages semantic features to control the representation of LLM's intermediate hidden states, enabling the model to meet specific requirements such as increased honesty or heightened safety awareness. However, a significant challenge arises when attempting to fulfill multiple requirements simultaneously. It proves difficult to encode various semantic contents, like honesty and safety, into a singular semantic feature, restricting its practicality. In this work, we address this challenge through Sparse Activation Control. By delving into the intrinsic mechanisms of LLMs, we manage to identify and pinpoint modules that are closely related to specific tasks within the model, i.e. attention heads. These heads display sparse characteristics that allow for near-independent control over different tasks. Our experiments, conducted on the open-source Llama series models, have yielded encouraging results. The models were able to align with human preferences on issues of safety, factualness, and bias concurrently.

While features of different scales are perceptually important to visual inputs, existing vision transformers do not yet take advantage of them explicitly. To this end, we first propose a cross-scale vision transformer, CrossFormer. It introduces a cross-scale embedding layer (CEL) and a long-short distance attention (LSDA). On the one hand, CEL blends each token with multiple patches of different scales, providing the self-attention module itself with cross-scale features. On the other hand, LSDA splits the self-attention module into a short-distance one and a long-distance counterpart, which not only reduces the computational burden but also keeps both small-scale and large-scale features in the tokens. Moreover, through experiments on CrossFormer, we observe another two issues that affect vision transformers' performance, i.e., the enlarging self-attention maps and amplitude explosion. Thus, we further propose a progressive group size (PGS) paradigm and an amplitude cooling layer (ACL) to alleviate the two issues, respectively. The CrossFormer incorporating with PGS and ACL is called CrossFormer++. Extensive experiments show that CrossFormer++ outperforms the other vision transformers on image classification, object detection, instance segmentation, and semantic segmentation tasks.

Previous monocular 3D detection works focus on the single frame input in both training and inference. In real-world applications, temporal and motion information naturally exists in monocular video. It is valuable for 3D detection but under-explored in monocular works. In this paper, we propose a straightforward and effective method for temporal feature fusion, which exhibits low computation cost and excellent transferability, making it conveniently applicable to various monocular models. Specifically, with the help of optical flow, we transform the backbone features produced by prior frames and fuse them into the current frame. We introduce the scene feature propagating mechanism, which accumulates history scene features without extra time-consuming. In this process, occluded areas are removed via forward-backward scene consistency. Our method naturally introduces valuable temporal features, facilitating 3D reasoning in monocular 3D detection. Furthermore, accumulated history scene features via scene propagating mitigate heavy computation overheads for video processing. Experiments are conducted on variant baselines, which demonstrate that the proposed method is model-agonistic and can bring significant improvement to multiple types of single-frame methods.

Graph Convolutional Networks (GCNs) suffer from performance degradation when models go deeper. However, earlier works only attributed the performance degeneration to over-smoothing. In this paper, we conduct theoretical and experimental analysis to explore the fundamental causes of performance degradation in deep GCNs: over-smoothing and gradient vanishing have a mutually reinforcing effect that causes the performance to deteriorate more quickly in deep GCNs. On the other hand, existing anti-over-smoothing methods all perform full convolutions up to the model depth. They could not well resist the exponential convergence of over-smoothing due to model depth increasing. In this work, we propose a simple yet effective plug-and-play module,SkipNode, to overcome the performance degradation of deep GCNs. It samples graph nodes in each convolutional layer to skip the convolution operation. In this way, both over-smoothing and gradient vanishing can be effectively suppressed since (1) not all nodes'features propagate through full layers and, (2) the gradient can be directly passed back through “skipped” nodes. We provide both theoretical analysis and empirical evaluation to demonstrate the efficacy ofSkipNodeand its superiority over SOTA baselines.

Deep neural network, despite its remarkable capability of discriminating targeted in-distribution samples, shows poor performance on detecting anomalous out-of-distribution data. To address this defect, state-of-the-art solutions choose to train deep networks on an auxiliary dataset of outliers. Various training criteria for these auxiliary outliers are proposed based on heuristic intuitions. However, we find that these intuitively designed outlier training criteria can hurt in-distribution learning and eventually lead to inferior performance. To this end, we identify three causes of the in-distribution incompatibility: contradictory gradient, false likelihood, and distribution shift. Based on our new understandings, we propose a new out-of-distribution detection method by adapting both the top-design of deep models and the loss function. Our method achieves in-distribution compatibility by pursuing less interference with the probabilistic characteristic of in-distribution features. On several benchmarks, our method not only achieves the state-of-the-art out-of-distribution detection performance but also improves the in-distribution accuracy.

Weakly supervised semantic segmentation (WSSS) with image-level labels is a challenging task. Mainstream approaches follow a multi-stage framework and suffer from high training costs. In this paper, we explore the potential of Contrastive Language-Image Pre-training models (CLIP) to localize different categories with only image-level labels and without further training. To efficiently generate high-quality segmentation masks from CLIP, we propose a novel WSSS framework called CLIP-ES. Our framework improves all three stages of WSSS with special designs for CLIP: 1) We introduce the softmax function into GradCAM and exploit the zero-shot ability of CLIP to suppress the confusion caused by non-target classes and backgrounds. Mean-while, to take full advantage of CLIP, we re-explore text inputs under the WSSS setting and customize two text-driven strategies: sharpness-based prompt selection and synonym fusion. 2) To simplify the stage of CAM refinement, we propose a real-time class-aware attention-based affinity (CAA) module based on the inherent multi-head self-attention (MHSA) in CLIP- ViTs. 3) When training the final segmentation model with the masks generated by CLIP, we introduced a confidence-guided loss (CGL) focus on confident regions. Our CLIP-ES achieves SOTA performance on Pascal VOC 2012 and MS COCO 2014 while only taking 10% time of previous methods for the pseudo mask generation. Code is available at https://github.com/linyq2117/CLIP-ES.

Recent Transformer-based 3D object detectors learn point cloud features either from point- or voxel-based representations. However, the former requires time-consuming sampling while the latter introduces quantization errors. In this paper, we present a novel Point-Voxel Transformer for single-stage 3D detection (PVT-SSD) that takes advantage of these two representations. Specifically, we first use voxel-based sparse convolutions for efficient feature encoding. Then, we propose a Point-Voxel Transformer (PVT) module that obtains long-range contexts in a cheap manner from voxels while attaining accurate positions from points. The key to associating the two different representations is our introduced input-dependent Query Initialization module, which could efficiently generate reference points and content queries. Then, PVT adaptively fuses long-range contextual and local geometric information around reference points into content queries. Further, to quickly find the neighboring points of reference points, we design the Virtual Range Image module, which generalizes the native range image to multi-sensor and multi-frame. The experiments on several autonomous driving benchmarks verify the effectiveness and efficiency of the proposed method. Code will be available.

Despite the tremendous progress of Masked Autoencoders (MAE) in developing vision tasks such as image and video, exploring MAE in large-scale 3D point clouds remains challenging due to the inherent irregularity. In contrast to previous 3D MAE frameworks, which either design a complex decoder to infer masked information from maintained regions or adopt sophisticated masking strategies, we instead propose a much simpler paradigm. The core idea is to apply a Generative Decoder for MAE (GD-MAE) to automatically merges the surrounding context to restore the masked geometric knowledge in a hierarchical fusion manner. In doing so, our approach is free from introducing the heuristic design of decoders and enjoys the flexibility of exploring various masking strategies. The corresponding part costs less than 12% latency compared with conventional methods, while achieving better performance. We demonstrate the efficacy of the proposed method on several large-scale benchmarks: Waymo, KITTI, and ONCE. Consistent improvement on downstream detection tasks illustrates strong robustness and generalization capability. Not only our method reveals state-of-the-art results, but remarkably, we achieve comparable accuracy even with 20% of the labeled data on the Waymo dataset. Code will be released.

Existing neural rendering methods for creating human avatars typically either require dense input signals such as video or multi-view images, or leverage a learned prior from large-scale specific 3D human datasets such that reconstruction can be performed with sparse-view inputs. Most of these methods fail to achieve realistic reconstruction when only a single image is available. To enable the data-efficient creation of realistic anima table 3D humans, we propose ELICIT, a novel method for learning human-specific neural radiance fields from a single image. Inspired by the fact that humans can effortlessly estimate the body geometry and imagine full-body clothing from a single image, we leverage two priors in ELICIT: 3D geometry prior and visual semantic prior. Specifically, ELICIT utilizes the 3D body shape geometry prior from a skinned vertex-based template model (i.e., SMPL) and implements the visual clothing semantic prior with the CLIP-based pre-trained models. Both priors are used to jointly guide the optimization for creating plausible content in the invisible areas. Taking advantage of the CLIP models, ELICIT can use text descriptions to generate text-conditioned unseen regions. In order to further improve visual details, we propose a segmentation-based sampling strategy that locally refines different parts of the avatar. Comprehensive evaluations on multiple popular benchmarks, including ZJU-MoCAP, Human3.6M, and DeepFashion, show that ELICIT outperforms strong baseline methods of avatar creation when only a single image is available. The code is public for research purposes at https://huangyangyi.github.io/ELICIT

Compared to typical multi-sensor systems, monocular 3D object detection has attracted much attention due to its simple configuration. However, there is still a significant gap between LiDAR-based and monocular-based methods. In this paper, we find that the ill-posed nature of monocular imagery can lead to depth ambiguity. Specifically, objects with different depths can appear with the same bounding boxes and similar visual features in the 2D image. Unfortunately, the network cannot accurately distinguish different depths from such non-discriminative visual features, resulting in unstable depth training. To facilitate depth learning, we propose a simple yet effective plug-and-play module, One Bounding Box Multiple Objects (OBMO). Concretely, we add a set of suitable pseudo labels by shifting the 3D bounding box along the viewing frustum. To constrain the pseudo-3D labels to be reasonable, we carefully design two label scoring strategies to represent their quality. In contrast to the original hard depth labels, such soft pseudo labels with quality scores allow the network to learn a reasonable depth range, boosting training stability and thus improving final performance. Extensive experiments on KITTI and Waymo benchmarks show that our method significantly improves state-of-the-art monocular 3D detectors by a significant margin (The improvements under the moderate setting on KITTI validation set are 1.82 ~ 10.91% mAP in BEV and 1.18 ~ 9.36% mAP in 3D). Codes have been released at https://github.com/mrsempress/OBMO.

Learning a distance function or metric on a given data manifold is of great importance in machine learning and pattern recognition. Many of the previous works first embed the manifold to Euclidean space and then learn the distance function. However, such a scheme might not faithfully preserve the distance function if the original manifold is not Euclidean. In this paper, we propose to learn the distance function directly on the manifold without embedding. We first provide a theoretical characterization of the distance function by its gradient field. Based on our theoretical analysis, we propose to first learn the gradient field of the distance function and then learn the distance function itself. Specifically, we set the gradient field of a local distance function as an initial vector field. Then we transport it to the whole manifold via heat flow on vector fields. Finally, the geodesic distance function can be obtained by requiring its gradient field to be close to the normalized vector field. Experimental results on both synthetic and real data demonstrate the effectiveness of our proposed algorithm.

Cross media retrieval systems have received increasing interest in recent years. Due to the semantic gap between low-level features and high-level semantic concepts of multimedia data, many researchers have explored joint-model techniques in cross media retrieval systems. Previous joint-model approaches usually focus on two traditional ways to design cross media retrieval systems: (a) fusing features from different media data; (b) learning different models for different media data and fusing their outputs. However, the process of fusing features or outputs will lose both low- and high-level abstraction information of media data. Hence, both ways do not really reveal the semantic correlations among the heterogeneous multimedia data. In this paper, we introduce a novel method for the cross media retrieval task, named Parallel Field Alignment Retrieval (PFAR), which integrates a manifold alignment framework from the perspective of vector fields. Instead of fusing original features or outputs, we consider the cross media retrieval as a manifold alignment problem using parallel fields. The proposed manifold alignment algorithm can effectively preserve the metric of data manifolds, model heterogeneous media data and project their relationship into intermediate latent semantic spaces during the process of manifold alignment. After the alignment, the semantic correlations are also determined. In this way, the cross media retrieval task can be resolved by the determined semantic correlations. Comprehensive experimental results have demonstrated the effectiveness of our approach.

Recently, ranking data with respect to the intrinsic geometric structure (manifold ranking) has received considerable attentions, with encouraging performance in many applications in pattern recognition, information retrieval and recommendation systems. Most of the existing manifold ranking methods focus on learning a ranking function that varies smoothly along the data manifold. However, beyond smoothness, a desirable ranking function should vary monotonically along the geodesics of the data manifold, such that the ranking order along the geodesics is preserved. In this article, we aim to learn a ranking function that varies linearly and therefore monotonically along the geodesics of the data manifold. Recent theoretical work shows that the gradient field of a linear function on the manifold has to be a parallel vector field. Therefore, we propose a novel ranking algorithm on the data manifolds, called Parallel Field Ranking. Specifically, we try to learn a ranking function and a vector field simultaneously. We require the vector field to be close to the gradient field of the ranking function, and the vector field to be as parallel as possible. Moreover, we require the value of the ranking function at the query point to be the highest, and then decrease linearly along the manifold. Experimental results on both synthetic data and real data demonstrate the effectiveness of our proposed algorithm.

Recently, ranking data with respect to the intrinsic geometric structure (manifold ranking) has received considerable attentions, with encouraging performance in many applications in pattern recognition, information retrieval and recommendation systems. Most of the existing manifold ranking methods focus on learning a ranking function that varies smoothly along the data manifold. However, beyond smoothness, a desirable ranking function should vary monotonically along the geodesics of the data manifold, such that the ranking order along the geodesics is preserved. In this paper, we aim to learn a ranking function that varies linearly and therefore monotonically along the geodesics of the data manifold. Recent theoretical work shows that the gradient field of a linear function on the manifold has to be a parallel vector field. Therefore, we propose a novel ranking algorithm on the data manifolds, called Parallel Field Ranking. Specifically, we try to learn a ranking function and a vector field simultaneously. We require the vector field to be close to the gradient field of the ranking function, and the vector field to be as parallel as possible. Moreover, we require the value of the ranking function at the query point to be the highest, and then decrease linearly along the manifold. Experimental results on both synthetic data and real data demonstrate the effectiveness of our proposed algorithm.

Multi-task learning (MTL) aims to improve generalization performance by learning multiple related tasks simultaneously and identifying the shared information among tasks. Most of existing MTL methods focus on learning linear models under the supervised setting. We propose a novel semi-supervised and nonlinear approach for MTL using vector fields. A vector field is a smooth mapping from the manifold to the tangent spaces which can be viewed as a directional derivative of functions on the manifold. We argue that vector fields provide a natural way to exploit the geometric structure of data as well as the shared differential structure of tasks, both are crucial for semi-supervised multi-task learning. In this paper, we develop multi-task vector field learning (MTVFL) which learns the prediction functions and the vector fields simultaneously. MTVFL has the following key properties: (1) the vector fields we learned are close to the gradient fields of the prediction functions; (2) within each task, the vector field is required to be as parallel as possible which is expected to span a low dimensional subspace; (3) the vector fields from all tasks share a low dimensional subspace. We formalize our idea in a regularization framework and also provide a convex relaxation method to solve the original non-convex problem. The experimental results on synthetic and real data demonstrate the effectiveness of our proposed approach.

This paper studies the problem of semi-supervised learning from the vector field perspective. Many of the existing work use the graph Laplacian to ensure the smoothness of the prediction function on the data manifold. However, beyond smoothness, it is suggested by recent theoretical work that we should ensure second order smoothness for achieving faster rates of convergence for semi-supervised regression problems. To achieve this goal, we show that the second order smoothness measures the linearity of the function, and the gradient field of a linear function has to be a parallel vector field. Consequently, we propose to find a function which minimizes the empirical error, and simultaneously requires its gradient field to be as parallel as possible. We give a continuous objective function on the manifold and discuss how to discretize it by using random points. The discretized optimization problem turns out to be a sparse linear system which can be solved very efficiently. The experimental results have demonstrated the effectiveness of our proposed approach.