Diffusion planning is a promising method for learning high-performance policies from offline data. To avoid the impact of discrepancies between planning and reality on performance, previous works generate new plans at each time step. However, this incurs significant computational overhead and leads to lower decision frequencies, and frequent plan switching may also affect performance. In contrast, humans might create detailed short-term plans and more general, sometimes vague, long-term plans, and adjust them over time. Inspired by this, we propose the Temporal Diffusion Planner (TDP) which improves decision efficiency by distributing the denoising steps across the time dimension. TDP begins by generating an initial plan that becomes progressively more vague over time. At each subsequent time step, rather than generating an entirely new plan, TDP updates the previous one with a small number of denoising steps. This reduces the average number of denoising steps, improving decision efficiency. Additionally, we introduce an automated replanning mechanism to prevent significant deviations between the plan and reality. Experiments on D4RL show that, compared to previous works that generate new plans every time step, TDP significantly improves the decision-making frequency by 11-24.8 times while achieving higher or comparable performance.

Large language models (LLMs) have shown promising capabilities in hardware description language (HDL) generation. However, existing approaches often rely on free-form natural language descriptions that are often ambiguous, redundant, and unstructured, which poses significant challenges for downstream Verilog code generation. We treat hardware code generation as a complex transformation from an open-ended natural language space to a domain-specific, highly constrained target space. To bridge this gap, we introduce Core Refined Understanding eXpression (CRUX), a structured intermediate space that captures the essential semantics of user intent while organizing the expression for precise Verilog code generation. We further design a two-stage training framework, comprising Joint Expression Modeling and Dual-Space Optimization, to enhance the quality of both CRUX and Verilog code. Experiments across multiple Verilog generation benchmarks demonstrate that our model, QiMeng-CRUX, achieves state-of-the-art performance among general models, particularly under challenging design tasks. Furthermore, the CRUX space proves transferable and beneficial when used as input prompts for other code models, highlighting its effectiveness in narrowing the gap between free-form natural language descriptions and precise Verilog generation.

Image restoration (IR) models are typically trained to recover high-quality images using L1 or LPIPS loss. To handle diverse unknown degradations, zero-shot IR methods have also been introduced. However, existing pre-trained and zero-shot IR approaches often fail to align with human preferences, resulting in restored images that may not be favored. This highlights the critical need to enhance restoration quality and adapt flexibly to various image restoration tasks or backbones without requiring model retraining and ideally without labor-intensive preference data collection. In this paper, we propose the first Test-Time Preference Optimization (TTPO) paradigm for image restoration, which enhances perceptual quality, generates preference data on-the-fly, and is compatible with any IR model backbone. Specifically, we design a training-free, three-stage pipeline: (i) generate candidate preference images online using diffusion inversion and denoising based on the initially restored image; (ii) select preferred and dispreferred images using automated preference-aligned metrics or human feedback; and (iii) use the selected preference images as reward signals to guide the diffusion denoising process, optimizing the restored image to better align with human preferences. Extensive experiments across various image restoration tasks and models demonstrate the effectiveness and flexibility of the proposed pipeline.

Vision-and-Language Navigation (VLN) requires an agent to dynamically explore complex 3D environments following human instructions. Recent research underscores the potential of harnessing large language models (LLMs) for VLN, given their commonsense knowledge and general reasoning capabilities. Despite their strengths, a substantial gap in task completion performance persists between LLM-based approaches and domain experts, as LLMs inherently struggle to comprehend real-world spatial correlations precisely; additionally, LLM inference can make the decision-making process considerably inefficient. To address these issues, we propose a novel dual-process thinking framework dubbed R3, integrating LLMs' generalization capabilities with VLN-specific expertise in a zero-shot manner. The framework comprises three core modules: Runner, Ruminator, and Regulator. The Runner is a lightweight transformer-based expert model that ensures efficient and accurate navigation under regular circumstances. The Ruminator employs a multimodal LLM as the backbone and adopts chain-of-thought (CoT) prompting to elicit structured reasoning from the LLM. The Regulator monitors the navigation progress and controls the appropriate thinking mode according to three criteria, integrating Runner and Ruminator harmoniously. Experimental results illustrate that R3 significantly outperforms other state-of-the-art methods, exceeding 3.28% and 3.30% in SPL and RGSPL respectively on the REVERIE benchmark, highlighting the effectiveness of our method in handling challenging VLN tasks.

Developing high-performance GPU kernels is critical for AI and scientific computing, but remains challenging due to its reliance on expert crafting and poor portability. While large language models (LLMs) offer promise for automation, both general-purpose and finetuned LLMs suffer from two fundamental and conflicting limitations: correctness and efficiency. The key reason is that existing LLM-based approaches directly generate the entire optimized low-level programs, requiring exploration of an extremely vast space encompassing both optimization policies and implementation codes. To address the challenge of exploring an intractable space, we propose Macro Thinking Micro Coding (MTMC), a hierarchical framework inspired by the staged optimization strategy of human experts. It decouples optimization strategy from implementation details, ensuring efficiency through high-level strategy and correctness through low-level implementation. Specifically, Macro Thinking employs reinforcement learning to guide lightweight LLMs in efficiently exploring and learning semantic optimization strategies that maximize hardware utilization. Micro Coding leverages general-purpose LLMs to incrementally implement the stepwise optimization proposals from Macro Thinking, avoiding full-kernel generation errors. Together, they effectively navigate the vast optimization space and intricate implementation details, enabling LLMs for high-performance GPU kernel generation. Comprehensive results on widely adopted benchmarks demonstrate the superior performance of MTMC on GPU kernel generation in both accuracy and running time. On KernelBench, MTMC achieves near 100% and 70% accuracy at Levels 1-2 and 3, over 50% than SOTA general-purpose and domain-finetuned LLMs, with up to 7.3× speedup over LLMs, and 2.2× over expert-optimized PyTorch Eager kernels. On the more challenging TritonBench, MTMC attains up to 59.64% accuracy and 34× speedup. All models and datasets will be made publicly available.

Changxin Ke, Rui Zhang, Jiaming Guo, Yuanbo Wen, Li Ding, Shuo Wang, Xuyuan Zhu, Xiong Peng, Di Huang, Zidong Du, Xing Hu, Qi Guo, Yunji Chen. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

3D Gaussian Splatting (3DGS) is a breakthrough in 3D reconstruction using 3D Gaussians. However, even on highend GPUs like the NVIDIA A100, reconstructing complex scenes remains time-consuming, taking over 15 minutes. The main bottleneck is α-computation, which accounts for 71.25 % of training workload, yet 93.03 % of it is invalid due to the localized influence of Gaussians. To address this issue, we propose Cambricon-GS, an accelerator for 3DGS training with Gaussian-Pixel hybrid parallelism. At the software level, we introduce a hybrid parallel workflow that breaks the limitation of conventional pixel-only parallelism through two key techniques: Center-Pixel Gaussian Culling (CPGC), which eliminates invalid Gaussians early, and SeedDriven Gaussian Region Exploration (SDGRE), which reduces invalid computation for partially valid Gaussians by selectively exploring valid regions. Overall, the workflow significantly reduces α computations, lowering the workload to 17.99 %. At the hardware level, Cambricon-GS decouples α-computation and α blending into GUnits and PUnits, organized in a 2D mesh-based NoC that supports asynchronous execution and efficient data routing. We further boost performance via Gaussian/Pixel load balancing and tiled SSIM based pipelining. The evaluation results show that Cambricon-GS achieves 19.63×, 14.86×, 15.42×, 2.98× and 2.63× speedup, and 78.62×, 63.00×, 61.72×, 3.89× and 3.22× energy saving, compared to A100, GSCore, GBU, GSArch, and GauSPU, respectively, with negligible image quality loss.

Hidden web interfaces in Internet of Things (IoT) devices pose significant security threats by unintentionally exposing inadequately protected functionalities, enabling attackers to bypass authentication, alter configurations, leak sensitive data, or execute arbitrary commands. Despite recent advancements, current detection approaches suffer from two critical challenges: 1) inadequately model the complex internal routing mechanisms of IoT firmware, leading to incomplete interface enumeration and substantial false negatives; and 2) inefficiently generate probing requests and verify unauthorized access due to limited semantic understanding of interface communication protocols. To overcome these challenges, we introduce FalconScope, a novel system combining precise firmware routing modeling and Large Language Model (LLM)-driven semantic analysis to detect hidden web interfaces effectively and efficiently. FalconScope achieves this through two key innovations: 1) a static analysis technique precisely reconstructs the device's internal routing mechanisms, enabling comprehensive enumeration of Routing Unique Identifiers and their corresponding backend handlers; and 2) an LLM-powered semantic engine automatically generates syntactically and semantically valid HTTP requests to efficiently trigger backend logic, coupled with semantic validation of device responses to accurately confirm unauthorized access. Evaluations on 11 real-world IoT devices from four major vendors demonstrate that FalconScope significantly surpasses existing state-of-the-art tools, detecting 620 hidden web interfaces—103 times more than IoTScope—while consuming only 3.6% of its analysis time. Following responsible disclosure, 50 issues have been assigned CVE IDs.

Diffusion Transformers (DiTs) have demonstrated remarkable success in text-to-video generation. However, the self-attention mechanism in DiTs imposes significant computational and memory burdens, particularly when handling long patch sequences like high-resolution or long-time videos. While sparse attention shows promise in reducing self-attention costs, existing approaches struggle to deliver performance gains due to the unique challenges in DiTs,i.e., varied sparse patterns across layers and timesteps, and the cumulative nature of inference errors over timesteps. In this paper, we propose DASA, an algorithm-hardware co-design that effectively addresses these challenges of attention sparsification in DiTs. Specifically, leveraging the insight that the generation quality is primarily influenced by overall distribution drift rather than changes in specific values, we introduce a novel Distribution-Aware Filtering (DAF) mechanism for sparsification. To further accelerate the process, we design a specialized Filtering Unit that enables fast candidate selection based on the proposed DAF mechanism. Experimental results show that DASA achieves 2.52× speed up compared to A100 GPU, and up to 1.22× speedup over state-of-the-art accelerators for self-attention computation.

We address the “long-range ambiguity” problem for unsupervised non-rigid point cloud correspondence, where corresponding points own inconsistent features while different local regions are spatially or geometrically similar. Previous methods struggle with this problem, since local reference frames (LRF) or coordinate-based methods struggle to exclude locally similar or spatially near mismatches, and widely used independent geometric relations might be inconsistent under non-rigid deformation, introducing extra ambiguity. To this end, we propose a novel robust context modeling module (RCM) to alleviatelong-range ambiguityin two aspects: 1) RCM tackles the ambiguity problem by introducing inter-relation attention (IRA), which mines robust cues from the interplay between relative geometric relations. 2) RCM enhances features with accessiblelong-rangeinformation from IRAs, following a local-to-global manner. Our method shows significant improvements in multiple benchmarks, with accurate correspondence over rotation and large deformation perturbation. Specifically, our method achieves a new state-of-the-art performance with correspondence accuracy of 33.9% and mean error of 4.2 on the SURREAL benchmark.

Odometry estimation remains a critical challenge for wheeled robots, as reducing its drift directly mitigates dependency on external localization systems. This paper proposes a distributed odometry framework for steerable wheels, named ICF-DO, which is applicable to both Steerable Wheeled Mobile Robots (SWMRs) and cooperative multi-single-wheel robot systems. The proposed method features low computational complexity and reduced drift, while demonstrating strong robustness in communication-restricted scenarios. Additionally, singularity can be processed in a distributed manner in the proposed framework. Experimental validation on a real physical SWMR platform demonstrates the effectiveness and practicality of the proposed method.

Neural Scene Representation (NSR) is a promising technique for representing real scenes.By learning from dozens of 2D photos captured from different viewpoints, NSR computes the 3D representation of real scenes.However, the performance of NSR processing running on GPU is insufficient for applications.Cambricon-R achieves high performance of more than 60 scenes per second, but at the cost of modeling quality.

Compilers, while essential, are notoriously complex systems that demand prohibitively expensive human expertise to develop and maintain. The recent advancements in Large Language Models (LLMs) offer a compelling new paradigm: Neural Compilation, which could potentially simplify compiler development for new architectures and facilitate the discovery of innovative optimization techniques. However, several critical obstacles impede its practical adoption. Firstly, a significant lack of dedicated benchmarks and robust evaluation methodologies hinders objective assessment and tracking of progress in the field. Secondly, systematically enhancing the reliability and performance of LLM-generated assembly remains a critical challenge. Addressing these challenges, this paper introduces NeuComBack, a novel benchmark dataset specifically designed for IR-to-assembly compilation. Leveraging this dataset, we first define a foundational Neural Compilation workflow and conduct a comprehensive evaluation of the capabilities of recent frontier LLMs on Neural Compilation, establishing new performance baselines. We further propose a self-evolving prompt optimization method that enables LLMs to iteratively evolve their internal prompt strategies by extracting insights from prior self-debugging traces, thereby enhancing their neural compilation capabilities. Experiments demonstrate that our method significantly improves both the functional correctness and the performance of LLM-generated assembly code. Compared to baseline prompts, the functional correctness rates improved from 44% to 64% on x86_64 and from 36% to 58% on aarch64, respectively. More significantly, among the 16 correctly generated x86_64 programs using our method, 14 (87.5%) surpassed clang-O3 performance. These consistent improvements across diverse architectures (x86_64 and aarch64) and program distributions (NeuComBack L1 and L2) validate our method's superiority over conventional approaches and its potential for broader adoption in low-level neural compilation.

The rise of GPU-based high-performance computing (HPC) has driven the widespread adoption of parallel programming models such as CUDA. Yet, the inherent complexity of parallel programming creates a demand for the automated sequential-to-parallel approaches. However, data scarcity poses a significant challenge for machine learning-based sequential-to-parallel code translation. Although recent back-translation methods show promise, they still fail to ensure functional equivalence in the translated code. In this paper, we propose \textbf{QiMeng-MuPa}, a novel \textbf{Mu}tual-Supervised Learning framework for Sequential-to-\textbf{Pa}rallel code translation, to address the functional equivalence issue. QiMeng-MuPa consists of two models, a Translator and a Tester. Through an iterative loop consisting of Co-verify and Co-evolve steps, the Translator and the Tester mutually generate data for each other and improve collectively. The Tester generates unit tests to verify and filter functionally equivalent translated code, thereby evolving the Translator, while the Translator generates translated code as augmented input to evolve the Tester. Experimental results demonstrate that QiMeng-MuPa significantly enhances the performance of the base models: when applied to Qwen2.5-Coder, it not only improves Pass@1 by up to 28.91\% and boosts Tester performance by 68.90\%, but also outperforms the previous state-of-the-art method CodeRosetta by 1.56 and 6.92 in BLEU and CodeBLEU scores, while achieving performance comparable to DeepSeek-R1 and GPT-4.1. Our code is available at \url{https://github.com/kcxain/mupa}.

The remarkable progress of Large Language Models (LLMs) presents promising opportunities for Verilog code generation which is significantly important for automated circuit design. The lacking of meaningful functional rewards hinders the preference optimization based on Reinforcement Learning (RL) for producing functionally correct Verilog code. In this paper, we propose Signal-Aware Learning for Verilog code generation (QiMeng-SALV) by leveraging code segments of functionally correct output signal to optimize RL training. Considering Verilog code specifies the structural interconnection of hardware gates and wires so that different output signals are independent, the key insight of QiMeng-SALV is to extract verified signal-aware implementations in partially incorrect modules, so as to enhance the extraction of meaningful functional rewards. Roughly, we verify the functional correctness of signals in generated module by comparing with that of reference module in the training data. Then abstract syntax tree (AST) is employed to identify signal-aware code segments which can provide meaningful functional rewards from erroneous modules. Finally, we introduce signal-aware DPO which is optimized on the correct signal-level code segments, thereby preventing noise and interference from incorrect signals. The proposed QiMeng-SALV underscores the paradigm shift from conventional module-level to fine-grained signal-level optimization in Verilog code generation, addressing the issue of insufficient functional rewards. Experiments demonstrate that our method achieves state-of-the-art performance on VerilogEval and RTLLM, with a 7B parameter model matching the performance of the DeepSeek v3 671B model and significantly outperforming the leading open-source model CodeV trained on the same dataset.

Dual-camera smartphones suffer from geometric and photometric inconsistencies during zoom transitions, primarily due to disparities in intrinsic/extrinsic parameters and divergent image processing pipelines between the two cameras. Existing interpolation methods struggle to effectively address this issue, constrained by the lack of ground-truth datasets and motion ambiguity in dynamic scenarios. To overcome these challenges, we propose OmniZoom, a universal plug-and-play paradigm for cross-device smooth zoom interpolation. Specifically, we present a novel cross-device virtual data generation method utilizing 3D Gaussian Splatting. This method tackles data scarcity by decoupling geometric features via spatial transition modeling and correcting photometric variations with dynamic color adaptation. It is further enhanced by cross-domain consistency learning for device-agnostic semantic alignment. Additionally, we introduce a plug-and-play 3D-TPR (3D Trajectory Progress Ratio Mapping) framework that surmounts 2D spatial limitations. As components of our framework, a texture-focus strategy is introduced for high-frequency detail preservation, incorporating mask penalty constraints to suppress interpolation artifacts. Our pipeline exhibits broad compatibility with diverse interpolation methods and achieves good performance across multiple public benchmarks. Real-world evaluations on various smartphone platforms also reveal significant quality improvements after fine-tuning on our synthetic data, which underscores the robustness and practical effectiveness of our approach for cross-device zoom applications.

Refactoring offers numerous advantages but also presents challenges such as high effort and limited automation. These challenges underscore the pressing need for an indepth analysis of past refactoring practices. However, existing empirical studies often focus on a limited set of refactoring types, narrow project scopes, or coarse-grained commit-level analysis, leaving several important questions underexplored. In particular, there is a lack of comprehensive understanding of how refactoring practices vary across ecosystems, evolve over time, and interact with external risks such as defect proneness and code instability. To address these gaps, this study presents an expanded empirical analysis of refactoring activities across diverse open-source projects. Leveraging an enhanced detection approach that captures nearly 100 types of refactorings, we conduct a multi-dimensional investigation into: the diversity of refactoring practices across ecosystems, their evolutionary patterns throughout the software lifecycle, and the association between bug proneness and post-refactoring volatility. Based on a comprehensive analysis of 807,651 refactoring operations, our findings provide a broader, lifecycle-aware and fine-grained perspective on how refactoring is applied in practice. We uncover consistent yet evolving usage patterns, reveal context-dependent risk profiles, and highlight the need for operation-level understanding of refactoring behavior. This study bridges key knowledge gaps in refactoring practices, offering an updated and richer understanding across three distinct dimensions and providing developers with empirical insights to optimize their refactoring strategies.

Large language models (LLMs) show their powerful automatic reasoning and planning capability with a wealth of semantic knowledge about the human world. However, the grounding problem still hinders the applications of LLMs in the real-world environment. Existing studies try to fine-tune the LLM or utilize pre-defined behavior APIs to bridge the LLMs and the environment, which not only costs huge human efforts to customize for every single task but also weakens the generality strengths of LLMs. To autonomously ground the LLM onto the environment, we proposed the Hypothesis, Verification, and Induction (HYVIN) framework to automatically and progressively ground the LLM with self-driven skill learning. HYVIN first employs the LLM to propose the hypothesis of sub-goals to achieve tasks and then verify the feasibility of the hypothesis via interacting with the underlying environment. Once verified, HYVIN can then learn generalized skills with the guidance of these successfully grounded subgoals. These skills can be further utilized to accomplish more complex tasks that fail to pass the verification phase. Verified in the famous instruction following task set, BabyAI, HYVIN achieves comparable performance in the most challenging tasks compared with imitation learning methods that cost millions of demonstrations, proving the effectiveness of learned skills and showing the feasibility and efficiency of our framework.

Model-based offline reinforcement learning (RL) algorithms have emerged as a promising paradigm for offline RL. These algorithms usually learn a dynamics model from a static dataset of transitions, use the model to generate synthetic trajectories, and perform conservative policy optimization within these trajectories. However, our observations indicate that policy optimization methods used in these model-based offline RL algorithms are not effective at exploring the learned model and induce biased exploration, which ultimately impairs the performance of the algorithm. To address this issue, we propose Offline Conservative ExplorAtioN (OCEAN), a novel rollout approach to model-based offline RL. In our method, we incorporate additional exploration techniques and introduce three conservative constraints based on uncertainty estimation to mitigate the potential impact of significant dynamic errors resulting from exploratory transitions. Our work is a plug-in method and can be combined with classical model-based RL algorithms, such as MOPO, COMBO, and RAMBO. Experiment results of our method on the D4RL MuJoCo benchmark show that OCEAN significantly improves the performance of existing algorithms.

Research on emergent communication has recently gained significant traction as a promising avenue for the linguistic community to unravel human language's origins and explore artificial intelligence's generalization capabilities. Current research has predominantly concentrated on recognizing qualitative patterns of object attributes(e.g., shape and color) and paid little attention to the quantitative relationship among object quantities which is known as the part of numerical concepts. The ability to generalize numerical concepts, i.e., counting and calculations with unseen quantities, is essential, as it mirrors humans' foundational abstract reasoning abilities. In this work, we introduce the NumGame, leveraging the referential game framework, forcing agents to communicate and generalize the numerical concepts effectively. Inspired by the human learning process of numbers, we present a two-stage training approach that sequentially fosters a rudimentary numerical sense followed by the ability of arithmetic calculation, ultimately aiding agents in generating semantically stable and unambiguous language for numerical concepts. The experimental results indicate the impressive generalization capabilities to unseen quantities and regularity of the language emergence from communication.

Huang Lei, Jiaming Guo, Guanhua He, Xishan Zhang, Rui Zhang, Shaohui Peng, Shaoli Liu, Tianshi Chen. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Overfitting in RL has become one of the main obstacles to applications in reinforcement learning(RL). Existing methods do not provide explicit semantic constrain for the feature extractor, hindering the agent from learning a unified cross-domain representation and resulting in performance degradation on unseen domains. Besides, abundant data from multiple domains are needed. To address these issues, in this work, we propose prompt-based visual alignment (PVA), a robust framework to mitigate the detrimental domain bias in the image for zero-shot policy transfer. Inspired that Visual-Language Model (VLM) can serve as a bridge to connect both text space and image space, we leverage the semantic information contained in a text sequence as an explicit constraint to train a visual aligner. Thus, the visual aligner can map images from multiple domains to a unified domain and achieve good generalization performance. To better depict semantic information, prompt tuning is applied to learn a sequence of learnable tokens. With explicit constraints of semantic information, PVA can learn unified cross-domain representation under limited access to cross-domain data and achieves great zero-shot generalization ability in unseen domains. We verify PVA on a vision-based autonomous driving task with CARLA simulator. Experiments show that the agent generalizes well on unseen domains under limited access to multi-domain data.

Despite the broad application of deep reinforcement learning (RL), transferring and adapting the policy to unseen but similar environments is still a significant challenge. Recently, the language-conditioned policy is proposed to facilitate policy transfer through learning the joint representation of observation and text that catches the compact and invariant information across various environments. Existing studies of language-conditioned RL methods often learn the joint representation as a simple latent layer for the given instances (episode-specific observation and text), which inevitably includes noisy or irrelevant information and cause spurious correlations that are dependent on instances, thus hurting generalization performance and training efficiency. To address the above issue, we propose a conceptual reinforcement learning (CRL) framework to learn the concept-like joint representation for language-conditioned policy. The key insight is that concepts are compact and invariant representations in human cognition through extracting similarities from numerous instances in real-world. In CRL, we propose a multi-level attention encoder and two mutual information constraints for learning compact and invariant concepts. Verified in two challenging environments, RTFM and Messenger, CRL significantly improves the training efficiency (up to 70%) and generalization ability (up to 30%) to the new environment dynamics.

Domain adaptation in RL mainly deals with the changes of observation when transferring the policy to a new environment. Many traditional approaches of domain adaptation in RL manage to learn a mapping function between the source and target domain in explicit or implicit ways. However, they typically require access to abundant data from the target domain. Besides, they often rely on visual clues to learn the mapping function and may fail when the source domain looks quite different from the target domain. To address these problems, in this paper, we propose a novel framework Online Prototype Alignment (OPA) to learn the mapping function based on the functional similarity of elements and is able to achieve few-shot policy transfer within only several episodes. The key insight of OPA is to introduce an exploration mechanism that can interact with the unseen elements of the target domain in an efficient and purposeful manner, and then connect them with the seen elements in the source domain according to their functionalities (instead of visual clues). Experimental results show that when the target domain looks visually different from the source domain, OPA can achieve better transfer performance even with much fewer samples from the target domain, outperforming prior methods.

Offline meta-reinforcement learning (OMRL) utilizes pre-collected offline datasets to enhance the agent's generalization ability on unseen tasks. However, the context shift problem arises due to the distribution discrepancy between the contexts used for training (from the behavior policy) and testing (from the exploration policy). The context shift problem leads to incorrect task inference and further deteriorates the generalization ability of the meta-policy. Existing OMRL methods either overlook this problem or attempt to mitigate it with additional information. In this paper, we propose a novel approach called Context Shift Reduction for OMRL (CSRO) to address the context shift problem with only offline datasets. The key insight of CSRO is to minimize the influence of policy in context during both the meta-training and meta-test phases. During meta-training, we design a max-min mutual information representation learning mechanism to diminish the impact of the behavior policy on task representation. In the meta-test phase, we introduce the non-prior context collection strategy to reduce the effect of the exploration policy. Experimental results demonstrate that CSRO significantly reduces the context shift and improves the generalization ability, surpassing previous methods across various challenging domains.

Deep reinforcement learning (DRL) has led to a wide range of advances in sequential decision-making tasks. However, the complexity of neural network policies makes it difficult to understand and deploy with limited computational resources. Currently, employing compact symbolic expressions as symbolic policies is a promising strategy to obtain simple and interpretable policies. Previous symbolic policy methods usually involve complex training processes and pre-trained neural network policies, which are inefficient and limit the application of symbolic policies. In this paper, we propose an efficient gradient-based learning method named Efficient Symbolic Policy Learning (ESPL) that learns the symbolic policy from scratch in an end-to-end way. We introduce a symbolic network as the search space and employ a path selector to find the compact symbolic policy. By doing so we represent the policy with a differentiable symbolic expression and train it in an off-policy manner which further improves the efficiency. In addition, in contrast with previous symbolic policies which only work in single-task RL because of complexity, we expand ESPL on meta-RL to generate symbolic policies for unseen tasks. Experimentally, we show that our approach generates symbolic policies with higher performance and greatly improves data efficiency for single-task RL. In meta-RL, we demonstrate that compared with neural network policies the proposed symbolic policy achieves higher performance and efficiency and shows the potential to be interpretable.

Research on emergent communication between deep-learning-based agents has received extensive attention due to its inspiration for linguistics and artificial intelligence. However, previous attempts have hovered around emerging communication under perception-oriented environmental settings, that forces agents to describe low-level perceptual features intra image or symbol contexts. In this work, inspired by the classic human reasoning test (namely Raven's Progressive Matrix), we propose the Reasoning Game, a cognition-oriented environment that encourages agents to reason and communicate high-level rules, rather than perceived low-level contexts. Moreover, we propose 1) an unbiased dataset (namely rule-RAVEN) as a benchmark to avoid overfitting, 2) and a two-stage curriculum agent training method as a baseline for more stable convergence in the Reasoning Game, where contexts and semantics are bilaterally drifting. Experimental results show that, in the Reasoning Game, a semantically stable and compositional language emerges to solve reasoning problems. The emerged language helps agents apply the extracted rules to the generalization of unseen context attributes, and to the transfer between different context attributes or even tasks.

In the field of multi-task reinforcement learning, the modular principle, which involves specializing functionalities into different modules and combining them appropriately, has been widely adopted as a promising approach to prevent the negative transfer problem that performance degradation due to conflicts between tasks. However, most of the existing multi-task RL methods only combine shared modules at the task level, ignoring that there may be conflicts within the task. In addition, these methods do not take into account that without constraints, some modules may learn similar functions, resulting in restricting the model's expressiveness and generalization capability of modular methods. In this paper, we propose the Contrastive Modules with Temporal Attention(CMTA) method to address these limitations. CMTA constrains the modules to be different from each other by contrastive learning and combining shared modules at a finer granularity than the task level with temporal attention, alleviating the negative transfer within the task and improving the generalization ability and the performance for multi-task RL. We conducted the experiment on Meta-World, a multi-task RL benchmark containing various robotics manipulation tasks. Experimental results show that CMTA outperforms learning each task individually for the first time and achieves substantial performance improvements over the baselines.

In recent years, Multi-Agent Reinforcement Learning (MARL) techniques have made significant strides in achieving high asymptotic performance in single task. However, there has been limited exploration of model transferability across tasks. Training a model from scratch for each task can be time-consuming and expensive, especially for large-scale Multi-Agent Systems. Therefore, it is crucial to develop methods for generalizing the model across tasks. Considering that there exist task-independent subtasks across MARL tasks, a model that can decompose such subtasks from the source task could generalize to target tasks. However, ensuring true task-independence of subtasks poses a challenge. In this paper, we propose to \textbf{d}ecompose a \textbf{t}ask in\textbf{to} a series of \textbf{g}eneralizable \textbf{s}ubtasks (DT2GS), a novel framework that addresses this challenge by utilizing a scalable subtask encoder and an adaptive subtask semantic module. We show that these components endow subtasks with two properties critical for task-independence: avoiding overfitting to the source task and maintaining consistent yet scalable semantics across tasks. Empirical results demonstrate that DT2GS possesses sound zero-shot generalization capability across tasks, exhibits sufficient transferability, and outperforms existing methods in both multi-task and single-task problems.

The majority of prevailing video interpolation methods compute flows to estimate the intermediate motion. However, accurate estimation of the intermediate motion is difficult with low-order motion model hypothesis, which induces enormous difficulties for subsequent processing. To alleviate the limitation, we propose a two-stage flow-free video interpolation architecture. Rather than utilizing pre-defined motion models, our method represents complex motion through data-driven learning. In the first stage, we analyze spatial-temporal information and generate coarse anchor frame features. In the second stage, we employ transformers to transfer neighboring features to the intermediate time steps and enhance the spatial textures. To improve the quality of coarse anchor frame features and the robustness in dealing with the multi-scale textures with large-scale motion, we propose a multi-scale architecture and transformers with variable token sizes to progressively enhance the features. The experimental results demonstrate that our model outperforms state-of-the-art methods for both single frame and multi frames interpolation tasks, and the extended ablation studies verify the effectiveness of our model.

Hierarchical reinforcement learning (HRL) has been proven to be effective for tasks with sparse rewards, for it can improve the agent's exploration efficiency by discovering high-quality hierarchical structures (e.g., subgoals or options). However, automatically discovering high-quality hierarchical structures is still a great challenge. Previous HRL methods can only find the hierarchical structures in simple environments, as they are mainly achieved through the randomness of agent's policies during exploration. In complicated environments, such a randomness-driven exploration paradigm can hardly discover high-quality hierarchical structures because of the low exploration efficiency. In this paper, we propose CDHRL, a causality-driven hierarchical reinforcement learning framework, to build high-quality hierarchical structures efficiently in complicated environments. The key insight is that the causalities among environment variables are naturally fit for modeling reachable subgoals and their dependencies; thus, the causality is suitable to be the guidance in building high-quality hierarchical structures. Roughly, we build the hierarchy of subgoals based on causality autonomously, and utilize the subgoal-based policies to unfold further causality efficiently. Therefore, CDHRL leverages a causality-driven discovery instead of a randomness-driven exploration for high-quality hierarchical structure construction. The results in two complex environments, 2D-Minecraft and Eden, show that CDHRL can discover high-quality hierarchical structures and significantly enhance exploration efficiency.

Object-oriented reinforcement learning (OORL) is a promising way to improve the sample efficiency and generalization ability over standard RL. Recent works that try to solve OORL tasks without additional feature engineering mainly focus on learning the object representations and then solving tasks via reasoning based on these object representations. However, none of these works tries to explicitly model the inherent similarity between different object instances of the same category. Objects of the same category should share similar functionalities; therefore, the category is the most critical property of an object. Following this insight, we propose a novel framework named Object-Category Aware Reinforcement Learning (OCARL), which utilizes the category information of objects to facilitate both perception and reasoning. OCARL consists of three parts: (1) Category-Aware Unsupervised Object Discovery (UOD), which discovers the objects as well as their corresponding categories; (2) Object-Category Aware Perception, which encodes the category information and is also robust to the incompleteness of (1) at the same time; (3) Object-Centric Modular Reasoning, which adopts multiple independent and object-category-specific networks when reasoning based on objects. Our experiments show that OCARL can improve both the sample efficiency and generalization in the OORL domain.

Policy gradient methods are appealing in deep reinforcement learning but suffer from high variance of gradient estimate. To reduce the variance, the state value function is applied commonly. However, the effect of the state value function becomes limited in stochastic dynamic environments, where the unexpected state dynamics and rewards will increase the variance. In this paper, we propose to replace the state value function with a novel hindsight value function, which leverages the information from the future to reduce the variance of the gradient estimate for stochastic dynamic environments. Particularly, to obtain an ideally unbiased gradient estimate, we propose an information-theoretic approach, which optimizes the embeddings of the future to be independent of previous actions. In our experiments, we apply the proposed hindsight value function in stochastic dynamic environments, including discrete-action environments and continuous-action environments. Compared with the standard state value function, the proposed hindsight value function consistently reduces the variance, stabilizes the training, and improves the eventual policy.

Winograd's minimal filtering algorithm has been widely used in Convolutional Neural Networks (CNNs) to reduce the number of multiplications for faster processing. However, it is only effective on convolutions with kernel size as 3x3 and stride as 1, because it suffers from significantly increased FLOPs and numerical accuracy problem for kernel size larger than 3x3 and fails on convolution with stride larger than 1. In this paper, we propose a novel Decomposable Winograd Method (DWM), which breaks through the limitation of original Winograd's minimal filtering algorithm to a wide and general convolutions. DWM decomposes kernels with large size or large stride to several small kernels with stride as 1 for further applying Winograd method, so that DWM can reduce the number of multiplications while keeping the numerical accuracy. It enables the fast exploring of larger kernel size and larger stride value in CNNs for high performance and accuracy and even the potential for new CNNs. Comparing against the original Winograd, the proposed DWM is able to support all kinds of convolutions with a speedup of ∼2, without affecting the numerical accuracy.

Cell detection and cell type classification from biomedical images play an important role for high-throughput imaging and various clinical application. While classification of single cell sample can be performed with standard computer vision and machine learning methods, analysis of multi-label samples (region containing congregating cells) is more challenging, as separation of individual cells can be difficult (e.g. touching cells) or even impossible (e.g. overlapping cells). As multi-instance images are common in analyzing Red Blood Cell (RBC) for Sickle Cell Disease (SCD) diagnosis, we develop and implement a multi-instance cell detection and classification framework to address this challenge. The framework firstly trains a region proposal model based on Region-based Convolutional Network (RCNN) to obtain bounding-boxes of regions potentially containing single or multiple cells from input microscopic images, which are extracted as image patches. High-level image features are then calculated from image patches through a pre-trained Convolutional Neural Network (CNN) with ResNet-50 structure. Using these image features inputs, six networks are then trained to make multi-label prediction of whether a given patch contains cells belonging to a specific cell type. As the six networks are trained with image patches consisting of both individual cells and touching/overlapping cells, they can effectively recognize cell types that are presented in multi-instance image samples. Finally, for the purpose of SCD testing, we train another machine learning classifier to predict whether the given image patch contains abnormal cell type based on outputs from the six networks. Testing result of the proposed framework shows that it can achieve good performance in automatic cell detection and classification.

Recent emerged quantization technique (i.e., using low bit-width fixed-point data instead of high bit-width floating-point data) has been applied to inference of deep neural networks for fast and efficient execution. However, directly applying quantization in training can cause significant accuracy loss, thus remaining an open challenge. In this paper, we propose a novel training approach, which applies a layer-wise precision-adaptive quantization in deep neural networks. The new training approach leverages our key insight that the degradation of training accuracy is attributed to the dramatic change of data distribution. Therefore, by keeping the data distribution stable through a layer-wise precision-adaptive quantization, we are able to directly train deep neural networks using low bit-width fixed-point data and achieve guaranteed accuracy, without changing hyper parameters. Experimental results on a wide variety of network architectures (e.g., convolution and recurrent networks) and applications (e.g., image classification, object detection, segmentation and machine translation) show that the proposed approach can train these neural networks with negligible accuracy losses (-1.40%-1.3%, 0.02% on average), and speed up training by 252% on a state-of-the-art Intel CPU.

Imaging-based early diagnosis of Alzheimer Disease (AD) has become an effective approach, especially by using nuclear medicine imaging techniques such as Positron Emission Topography (PET). In various literature it has been found that PET images can be better modeled as signals (e.g. uptake of florbetapir) defined on a network (non-Euclidean) structure which is governed by its underlying graph patterns of pathological progression and metabolic connectivity. In order to effectively apply deep learning framework for PET image analysis to overcome its limitation on Euclidean grid, we develop a solution for 3D PET image representation and analysis under a generalized, graph-based CNN architecture (PETNet), which analyzes PET signals defined on a group-wise inferred graph structure. Computations in PETNet are defined in non-Euclidean, graph (network) domain, as it performs feature extraction by convolution operations on spectral-filtered signals on the graph and pooling operations based on hierarchical graph clustering. Effectiveness of the PETNet is evaluated on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset, which shows improved performance over both deep learning and other machine learning-based methods.

Virtual try-on system could demonstrate the visual effect of wearing certain clothes, which is in a great demand for the online clothing customers. Due to the neglect of the original person structure, the previous try-on schemes usually encounter with the problems like the loss of body parts, the missing of body details and the deviation of clothing style. In this paper, we propose a novel image-based virtual try-on network, which could maintain the structural consistency between the generated image and the original image by human parsing. Our network consists of three components. Given the original person and the clothing images, the target human parsing maps are generated. Then, the parsing maps are matched with the target clothes to generate the warped clothes. Finally, according to the parsing results, the parts to be replaced the original images are intercepted, and more original information is retained as the input of the network to generate the final results. Experiments on an existing benchmark demonstrate our method maintains the consistency of structure and achieves the state-of-the-art performance.

Clothing parsing is significant to many clothing applications. Recently, a lot of clothing parsing methods have been presented, which explore the innovation of the parsing pipeline or try to find more specific prior information. Although these methods perform well in some benchmarks, a few challenging problems have not been solved yet, such as the complicated mutual interference among labels. In this study, the authors propose a Conditional Progressive Network to parse clothing in different scales and prevent the mutual interference among labels. The authors’ solution consists of three sub‐networks, including Conditional Parsing Network (CPN), Pose Estimation Network (PEN) and Label Transform Network (LTN). Specifically, the CPN module generates the intermediate parsing result in the form of the multiple progressive stages, which combines with the previous outputs in each stage and the specific prior conditions. The PEN module provides a series of heat maps about the human pose information. The LTN module suppresses the redundant labels to avoid the mutual interference among labels. They demonstrate their solution in parsing the fashion clothing cases on the ATR and the Fashion dataset. In their experiments, their method obtains a better performance than the state‐of‐the‐art methods.

Human parsing, which segments a human-centric image into pixel-wise categorization, has a wide range of applications. However, none of the existing methods can productively solve the issue of label parsing fragmentation due to confused and complicated annotations. In this paper, we propose a novel Trusted Guidance Pyramid Network (TGPNet) to address this limitation. Based on a pyramid architecture, we design a Pyramid Residual Pooling (PRP) module setting at the end of a bottom-up approach to capture both global and local level context. In the top-down approach, we propose a Trusted Guidance Multi-scale Supervision (TGMS) that efficiently integrates and supervises multi-scale contextual information. Furthermore, we present a simple yet powerful Trusted Guidance Framework (TGF) which imposes global-level semantics into parsing results directly without extra ground truth labels in model training. Extensive experiments on two public human parsing benchmarks well demonstrate that our TGPNet has a strong ability in solving label parsing fragmentation problem and has an obtained improvement than other methods.

Extracting an object of interest from a single video still faces significant difficulties when the object has variegated appearance, manifests articulated motion, or experiences occlusions by other objects. In this paper, we present a video cosegmentation method to address the aforementioned challenges. Departing from the objectness attributes and motion coherence used by traditional foreground-background separation and video segmentation methods, we place central importance in the role of “common fate.” Specifically, the different parts of the object should persist together in all the videos despite the possible presence of incoherent (e.g., articulated) motions. To accomplish this idea, we first extract seed superpixels by a motion-based foreground segmentation method. We next formulate a set of initial to-link constraints between these superpixels based on whether they exhibit the characteristics of common fate. An iterative manifold ranking algorithm is then proposed to trim away the incorrect and accidental linkage relationships. Having discovered the parts that should cohere together, we next perform clustering to extract the entire object and to handle the case in which there might be multiple objects present. This clustering is performed at two levels: the superpixel level and the object level. This two-level clustering algorithm also performs automatic model selection to estimate the number of object classes extracted. Finally, a multiclass labeling Markov random field is used to obtain a refined segmentation result. To evaluate the performance of our framework, we introduce a new data set in which the videos have complex form and motion that are liable to ambiguity in interpretation. Our experimental results on this data set show that our method successfully addresses the challenges in the extraction of complex objects and outperforms the state-of-the-art video segmentation and cosegmentation methods in our data set.

Given a pair of videos having a common action, our goal is to simultaneously segment this pair of videos to extract this common action. As a preprocessing step, we first remove background trajectories by a motion-based figure ground segmentation. To remove the remaining background and those extraneous actions, we propose the trajectory co saliency measure, which captures the notion that trajectories recurring in all the videos should have their mutual saliency boosted. This requires a trajectory matching process which can compare trajectories with different lengths and not necessarily spatiotemporally aligned, and yet be discriminative enough despite significant intra-class variation in the common action. We further leverage the graph matching to enforce geometric coherence between regions so as to reduce feature ambiguity and matching errors. Finally, to classify the trajectories into common action and action outliers, we formulate the problem as a binary labeling of a Markov Random Field, in which the data term is measured by the trajectory co-saliency and the smoothness term is measured by the spatiotemporal consistency between trajectories. To evaluate the performance of our framework, we introduce a dataset containing clips that have animal actions as well as human actions. Experimental results show that the proposed method performs well in common action extraction.

This paper addresses real-world challenges in the motion segmentation problem, including perspective effects, missing data, and unknown number of motions. It first formulates the 3-D motion segmentation from two perspective views as a subspace clustering problem, utilizing the epipolar constraint of an image pair. It then combines the point correspondence information across multiple image frames via a collaborative clustering step, in which tight integration is achieved via a mixed norm optimization scheme. For model selection, we propose an over-segment and merge approach, where the merging step is based on the property of the ell_1-norm of the mutual sparse representation of two over-segmented groups. The resulting algorithm can deal with incomplete trajectories and perspective effects substantially better than state-of-the-art two-frame and multi-frame methods. Experiments on a 62-clip dataset show the significant superiority of the proposed idea in both segmentation accuracy and model selection.

This paper presents a new system level framework for Diminished Reality, leveraging for the first time both the appearance and 3D information provided by large photo collections on the Internet. Recent computer vision techniques have made it possible to automatically reconstruct 3-D structure-from-motion points from large and unordered photo collections. Using these point clouds and a prior provided by GPS, reasonably accurate 6 degree of freedom camera poses can be obtained, thus allowing localization. Once the camera (and hence the user) is correctly localized, photos depicting scenes visible from the user's viewpoint can be used to remove unwanted objects indicated by the user in the video sequences. Existing methods based on texture synthesis bring undesirable artifacts and video inconsistency when the background is heterogeneous; the task is rendered even harder for these methods when the background contains complex structures. On the other hand, methods based on plane warping fail when the background has arbitrary shape. Unlike these methods, our algorithm copes with these problems by making use of internet photos, registering them in 3D space and obtaining the 3D scene structure in an offline process. We carefully design the various components during the online phase so as to meet both speed and quality requirements of the task. Experiments on real data collected demonstrate the superiority of our system.

In this paper, we propose a novel method to generate the cyclostationary signature for Orthogonal Frequency Division Multiplexing (OFDM) signals with a low overhead of signature (i.e. number of signature subcarriers) and a good performance of the signature detection under frequency-selective fading channel, for dynamic spectrum access (DSA) networks. Moreover, we also present how to employ the proposed method to detect the embedded signature. Simulations are conducted to make a comparison of the detection performance between the proposed method and the Sutton method. Relying on single hopping subcarrier to generate the signature, the overhead of the proposed method is only one fourth that of the Sutton method in our example. Nevertheless, the simulation results showed that the detection performance of the proposed method is no worse than that of the method proposed by Sutton.

In this paper, we propose an approach of multi-antenna spectrum sensing based on cyclostationarity for cognitive radio. The key idea of the proposed method is to extend Dandawate's generalized likelihood ratio test to take into account the estimation of all cyclic cross-correlation as well as all cyclic autocorrelation obtainable in a multi-antenna system. The proposed method is able to take advantage of spatial diversity without any prior knowledge or estimation of channel information. Furthermore, we propose a simplified method to replace the full generalized likelihood ratio test in order to reduce the computational complexity. Simulation results show the reliability of our proposed detector and demonstrate the effectiveness of using the cyclic cross-correlations, which contribute to a performance gain of approximately 2dB when a four-antenna receiver is considered.