Wei-Shi Zheng

Xiao-Ming Wu, Bin Fan, Kang Liao et al.

Following the rise of large foundation models, Vision-Language-Action models (VLAs) emerged, leveraging strong visual and language understanding from Vision-Language Models for general-purpose policy learning. Yet, the current VLA landscape remains fragmented and exploratory. Although many groups have proposed their own VLA models, inconsistencies in training protocols and evaluation settings make it difficult to identify which design choices truly matter. To bring structure to this evolving space, we reexamine the VLA design space under a unified framework and evaluation setup. Starting from a simple VLA baseline similar to RT-2, which is the origin of VLA, we systematically dissect design choices along three dimensions: foundational components, perception essentials, and action modelling perspectives. From this study, we distill 12 key findings that together form a practical recipe for building strong VLA models. The outcome of this exploration is a simple yet effective model, VLANeXt. It outperforms the state-of-the-art methods on the LIBERO and LIBERO-plus benchmarks and demonstrates strong performance in real-world experiments. We release a unified and easy-to-use codebase to reproduce our findings, explore the design space, and develop new VLA variants on top of a shared foundation. The codebase is available at https://github.com/DravenALG/VLANeXt.

Chengzhi Lin, Ancong Wu, Junwei Liang et al. · cmu, tencent-ai

Cross-modal retrieval between videos and texts has gained increasing research interest due to the rapid emergence of videos on the web. Generally, a video contains rich instance and event information and the query text only describes a part of the information. Thus, a video can correspond to multiple different text descriptions and queries. We call this phenomenon the ``Video-Text Correspondence Ambiguity'' problem. Current techniques mostly concentrate on mining local or multi-level alignment between contents of a video and text (\textit{e.g.}, object to entity and action to verb). It is difficult for these methods to alleviate the video-text correspondence ambiguity by describing a video using only one single feature, which is required to be matched with multiple different text features at the same time. To address this problem, we propose a Text-Adaptive Multiple Visual Prototype Matching model, which automatically captures multiple prototypes to describe a video by adaptive aggregation of video token features. Given a query text, the similarity is determined by the most similar prototype to find correspondence in the video, which is termed text-adaptive matching. To learn diverse prototypes for representing the rich information in videos, we propose a variance loss to encourage different prototypes to attend to different contents of the video. Our method outperforms state-of-the-art methods on four public video retrieval datasets.

Gaojie Wu, Wei-Shi Zheng, Yutong Lu et al.

Vision Transformer (ViT) has shown great potential for various visual tasks due to its ability to model long-range dependency. However, ViT requires a large amount of computing resource to compute the global self-attention. In this work, we propose a ladder self-attention block with multiple branches and a progressive shift mechanism to develop a light-weight transformer backbone that requires less computing resources (e.g. a relatively small number of parameters and FLOPs), termed Progressive Shift Ladder Transformer (PSLT). First, the ladder self-attention block reduces the computational cost by modelling local self-attention in each branch. In the meanwhile, the progressive shift mechanism is proposed to enlarge the receptive field in the ladder self-attention block by modelling diverse local self-attention for each branch and interacting among these branches. Second, the input feature of the ladder self-attention block is split equally along the channel dimension for each branch, which considerably reduces the computational cost in the ladder self-attention block (with nearly 1/3 the amount of parameters and FLOPs), and the outputs of these branches are then collaborated by a pixel-adaptive fusion. Therefore, the ladder self-attention block with a relatively small number of parameters and FLOPs is capable of modelling long-range interactions. Based on the ladder self-attention block, PSLT performs well on several vision tasks, including image classification, objection detection and person re-identification. On the ImageNet-1k dataset, PSLT achieves a top-1 accuracy of 79.9% with 9.2M parameters and 1.9G FLOPs, which is comparable to several existing models with more than 20M parameters and 4G FLOPs. Code is available at https://isee-ai.cn/wugaojie/PSLT.html.

Yu-Ming Tang, Yi-Xing Peng, Wei-Shi Zheng

Incremental learning aims to overcome catastrophic forgetting when learning deep networks from sequential tasks. With impressive learning efficiency and performance, prompt-based methods adopt a fixed backbone to sequential tasks by learning task-specific prompts. However, existing prompt-based methods heavily rely on strong pretraining (typically trained on ImageNet-21k), and we find that their models could be trapped if the potential gap between the pretraining task and unknown future tasks is large. In this work, we develop a learnable Adaptive Prompt Generator (APG). The key is to unify the prompt retrieval and prompt learning processes into a learnable prompt generator. Hence, the whole prompting process can be optimized to reduce the negative effects of the gap between tasks effectively. To make our APG avoid learning ineffective knowledge, we maintain a knowledge pool to regularize APG with the feature distribution of each class. Extensive experiments show that our method significantly outperforms advanced methods in exemplar-free incremental learning without (strong) pretraining. Besides, under strong retraining, our method also has comparable performance to existing prompt-based models, showing that our method can still benefit from pretraining. Codes can be found at https://github.com/TOM-tym/APG

Yipeng Gao, Zeyu Wang, Wei-Shi Zheng et al.

Contrastive learning has emerged as a promising paradigm for 3D open-world understanding, i.e., aligning point cloud representation to image and text embedding space individually. In this paper, we introduce MixCon3D, a simple yet effective method aiming to sculpt holistic 3D representation in contrastive language-image-3D pre-training. In contrast to point cloud only, we develop the 3D object-level representation from complementary perspectives, e.g., multi-view rendered images with the point cloud. Then, MixCon3D performs language-3D contrastive learning, comprehensively depicting real-world 3D objects and bolstering text alignment. Additionally, we pioneer the first thorough investigation of various training recipes for the 3D contrastive learning paradigm, building a solid baseline with improved performance. Extensive experiments conducted on three representative benchmarks reveal that our method significantly improves over the baseline, surpassing the previous state-of-the-art performance on the challenging 1,156-category Objaverse-LVIS dataset by 5.7%. The versatility of MixCon3D is showcased in applications such as text-to-3D retrieval and point cloud captioning, further evidencing its efficacy in diverse scenarios. The code is available at https://github.com/UCSC-VLAA/MixCon3D.

Shuxuan Li, Zhilin Zhao, Quyu Kong et al.

Performance estimation under distribution shift aims to predict how a model behaves on an unlabeled test set whose distribution differs from the training data, a scenario that requires reliable indicators that can faithfully reflect model behavior without ground-truth labels. Existing approaches rely solely on the outputs of the given model whose biases are amplified once the distribution shifts, weakening the correlation with the true performance. Motivated by this limitation, we propose Fused Reference Alignment Prediction (FRAP), which leverages the complementary strengths of an external foundation model and the base model to construct a more reliable surrogate of the ground-truth labels. FRAP aligns the prediction distribution of the foundation model with that of the base model by applying temperature-scaled calibration that minimizes their divergence. The aligned predictions are fused through confidence-based weighting into a refined reference distribution that integrates robustness from the foundation model and domain-specific expertise from the base model, and performance estimation is obtained by measuring how closely the base model predictions agree with this reference. Extensive experiments across diverse datasets and architectures show that FRAP provides consistent and substantial improvements over representative performance-estimation methods under distribution shift.

Yaojia Zheng, Zhouwu Liu, Rong Mo et al.

Accurate automated analysis of electroencephalography (EEG) would largely help clinicians effectively monitor and diagnose patients with various brain diseases. Compared to supervised learning with labelled disease EEG data which can train a model to analyze specific diseases but would fail to monitor previously unseen statuses, anomaly detection based on only normal EEGs can detect any potential anomaly in new EEGs. Different from existing anomaly detection strategies which do not consider any property of unavailable abnormal data during model development, a task-oriented self-supervised learning approach is proposed here which makes use of available normal EEGs and expert knowledge about abnormal EEGs to train a more effective feature extractor for the subsequent development of anomaly detector. In addition, a specific two branch convolutional neural network with larger kernels is designed as the feature extractor such that it can more easily extract both larger scale and small-scale features which often appear in unavailable abnormal EEGs. The effectively designed and trained feature extractor has shown to be able to extract better feature representations from EEGs for development of anomaly detector based on normal data and future anomaly detection for new EEGs, as demonstrated on three EEG datasets. The code is available at https://github.com/ironing/EEG-AD.

Qijie Mo, Yipeng Gao, Shenghao Fu et al.

In incremental object detection, knowledge distillation has been proven to be an effective way to alleviate catastrophic forgetting. However, previous works focused on preserving the knowledge of old models, ignoring that images could simultaneously contain categories from past, present, and future stages. The co-occurrence of objects makes the optimization objectives inconsistent across different stages since the definition for foreground objects differs across various stages, which limits the model's performance greatly. To overcome this problem, we propose a method called ``Bridge Past and Future'' (BPF), which aligns models across stages, ensuring consistent optimization directions. In addition, we propose a novel Distillation with Future (DwF) loss, fully leveraging the background probability to mitigate the forgetting of old classes while ensuring a high level of adaptability in learning new classes. Extensive experiments are conducted on both Pascal VOC and MS COCO benchmarks. Without memory, BPF outperforms current state-of-the-art methods under various settings. The code is available at https://github.com/iSEE-Laboratory/BPF.

Shenghao Fu, Junkai Yan, Yipeng Gao et al.

Recent sparse detectors with multiple, e.g. six, decoder layers achieve promising performance but much inference time due to complex heads. Previous works have explored using dense priors as initialization and built one-decoder-layer detectors. Although they gain remarkable acceleration, their performance still lags behind their six-decoder-layer counterparts by a large margin. In this work, we aim to bridge this performance gap while retaining fast speed. We find that the architecture discrepancy between dense and sparse detectors leads to feature conflict, hampering the performance of one-decoder-layer detectors. Thus we propose Adaptive Sparse Anchor Generator (ASAG) which predicts dynamic anchors on patches rather than grids in a sparse way so that it alleviates the feature conflict problem. For each image, ASAG dynamically selects which feature maps and which locations to predict, forming a fully adaptive way to generate image-specific anchors. Further, a simple and effective Query Weighting method eases the training instability from adaptiveness. Extensive experiments show that our method outperforms dense-initialized ones and achieves a better speed-accuracy trade-off. The code is available at \url{https://github.com/iSEE-Laboratory/ASAG}.

Jia-Run Du, Jia-Chang Feng, Kun-Yu Lin et al. · tencent-ai

Weakly Supervised Temporal Action Localization (WSTAL) aims to localize and classify action instances in long untrimmed videos with only video-level category labels. Due to the lack of snippet-level supervision for indicating action boundaries, previous methods typically assign pseudo labels for unlabeled snippets. However, since some action instances of different categories are visually similar, it is non-trivial to exactly label the (usually) one action category for a snippet, and incorrect pseudo labels would impair the localization performance. To address this problem, we propose a novel method from a category exclusion perspective, named Progressive Complementary Learning (ProCL), which gradually enhances the snippet-level supervision. Our method is inspired by the fact that video-level labels precisely indicate the categories that all snippets surely do not belong to, which is ignored by previous works. Accordingly, we first exclude these surely non-existent categories by a complementary learning loss. And then, we introduce the background-aware pseudo complementary labeling in order to exclude more categories for snippets of less ambiguity. Furthermore, for the remaining ambiguous snippets, we attempt to reduce the ambiguity by distinguishing foreground actions from the background. Extensive experimental results show that our method achieves new state-of-the-art performance on two popular benchmarks, namely THUMOS14 and ActivityNet1.3.

Delong Zhang, Yi-Xing Peng, Xiao-Ming Wu et al.

Online person re-identification services face privacy breaches from potential data leakage and recovery attacks, exposing cloud-stored images to malicious attackers and triggering public concern. The privacy protection of pedestrian images is crucial. Previous privacy-preserving person re-identification methods are unable to resist recovery attacks and compromise accuracy. In this paper, we propose an iterative method (PixelFade) to optimize pedestrian images into noise-like images to resist recovery attacks. We first give an in-depth study of protected images from previous privacy methods, which reveal that the chaos of protected images can disrupt the learning of recovery models. Accordingly, Specifically, we propose Noise-guided Objective Function with the feature constraints of a specific authorization model, optimizing pedestrian images to normal-distributed noise images while preserving their original identity information as per the authorization model. To solve the above non-convex optimization problem, we propose a heuristic optimization algorithm that alternately performs the Constraint Operation and the Partial Replacement Operation. This strategy not only safeguards that original pixels are replaced with noises to protect privacy, but also guides the images towards an improved optimization direction to effectively preserve discriminative features. Extensive experiments demonstrate that our PixelFade outperforms previous methods in resisting recovery attacks and Re-ID performance. The code is available at https://github.com/iSEE-Laboratory/PixelFade.

Yang Yang, Zhiying Cui, Junjie Xu et al.

Deep learning has shown its human-level performance in various applications. However, current deep learning models are characterised by catastrophic forgetting of old knowledge when learning new classes. This poses a challenge particularly in intelligent diagnosis systems where initially only training data of a limited number of diseases are available. In this case, updating the intelligent system with data of new diseases would inevitably downgrade its performance on previously learned diseases. Inspired by the process of learning new knowledge in human brains, we propose a Bayesian generative model for continual learning built on a fixed pre-trained feature extractor. In this model, knowledge of each old class can be compactly represented by a collection of statistical distributions, e.g. with Gaussian mixture models, and naturally kept from forgetting in continual learning over time. Unlike existing class-incremental learning methods, the proposed approach is not sensitive to the continual learning process and can be additionally well applied to the data-incremental learning scenario. Experiments on multiple medical and natural image classification tasks showed that the proposed approach outperforms state-of-the-art approaches which even keep some images of old classes during continual learning of new classes.

Yu-Ming Tang, Yi-Xing Peng, Jingke Meng et al.

Few-shot class-incremental learning (FSCIL) aims to learn sequential classes with limited samples in a few-shot fashion. Inherited from the classical class-incremental learning setting, the popular benchmark of FSCIL uses averaged accuracy (aAcc) and last-task averaged accuracy (lAcc) as the evaluation metrics. However, we reveal that such evaluation metrics may not provide adequate emphasis on the novel class performance, and the continual learning ability of FSCIL methods could be ignored under this benchmark. In this work, as a complement to existing metrics, we offer a new metric called generalized average accuracy (gAcc) which is designed to provide an extra equitable evaluation by incorporating different perspectives of the performance under the guidance of a parameter $α$. We also present an overall metric in the form of the area under the curve (AUC) along the $α$. Under the guidance of gAcc, we release the potential of intermediate features of the vision transformers to boost the novel-class performance. Taking information from intermediate layers which are less class-specific and more generalizable, we manage to rectify the final features, leading to a more generalizable transformer-based FSCIL framework. Without complex network designs or cumbersome training procedures, our method outperforms existing FSCIL methods at aAcc and gAcc on three datasets. See codes at https://github.com/iSEE-Laboratory/Revisting_FSCIL

Yuan-Ming Li, Ling-An Zeng, Jing-Ke Meng et al.

Action Quality Assessment (AQA) is a task that tries to answer how well an action is carried out. While remarkable progress has been achieved, existing works on AQA assume that all the training data are visible for training at one time, but do not enable continual learning on assessing new technical actions. In this work, we address such a Continual Learning problem in AQA (Continual-AQA), which urges a unified model to learn AQA tasks sequentially without forgetting. Our idea for modeling Continual-AQA is to sequentially learn a task-consistent score-discriminative feature distribution, in which the latent features express a strong correlation with the score labels regardless of the task or action types.From this perspective, we aim to mitigate the forgetting in Continual-AQA from two aspects. Firstly, to fuse the features of new and previous data into a score-discriminative distribution, a novel Feature-Score Correlation-Aware Rehearsal is proposed to store and reuse data from previous tasks with limited memory size. Secondly, an Action General-Specific Graph is developed to learn and decouple the action-general and action-specific knowledge so that the task-consistent score-discriminative features can be better extracted across various tasks. Extensive experiments are conducted to evaluate the contributions of proposed components. The comparisons with the existing continual learning methods additionally verify the effectiveness and versatility of our approach. Data and code are available at https://github.com/iSEE-Laboratory/Continual-AQA.

Xiao-Ming Wu, Dian Zheng, Zuhao Liu et al.

Binarization of neural networks is a dominant paradigm in neural networks compression. The pioneering work BinaryConnect uses Straight Through Estimator (STE) to mimic the gradients of the sign function, but it also causes the crucial inconsistency problem. Most of the previous methods design different estimators instead of STE to mitigate it. However, they ignore the fact that when reducing the estimating error, the gradient stability will decrease concomitantly. These highly divergent gradients will harm the model training and increase the risk of gradient vanishing and gradient exploding. To fully take the gradient stability into consideration, we present a new perspective to the BNNs training, regarding it as the equilibrium between the estimating error and the gradient stability. In this view, we firstly design two indicators to quantitatively demonstrate the equilibrium phenomenon. In addition, in order to balance the estimating error and the gradient stability well, we revise the original straight through estimator and propose a power function based estimator, Rectified Straight Through Estimator (ReSTE for short). Comparing to other estimators, ReSTE is rational and capable of flexibly balancing the estimating error with the gradient stability. Extensive experiments on CIFAR-10 and ImageNet datasets show that ReSTE has excellent performance and surpasses the state-of-the-art methods without any auxiliary modules or losses.

Xiao-Ming Wu, Jia-Feng Cai, Jian-Jian Jiang et al.

Robotic grasping in clutters is a fundamental task in robotic manipulation. In this work, we propose an economic framework for 6-DoF grasp detection, aiming to economize the resource cost in training and meanwhile maintain effective grasp performance. To begin with, we discover that the dense supervision is the bottleneck of current SOTA methods that severely encumbers the entire training overload, meanwhile making the training difficult to converge. To solve the above problem, we first propose an economic supervision paradigm for efficient and effective grasping. This paradigm includes a well-designed supervision selection strategy, selecting key labels basically without ambiguity, and an economic pipeline to enable the training after selection. Furthermore, benefit from the economic supervision, we can focus on a specific grasp, and thus we devise a focal representation module, which comprises an interactive grasp head and a composite score estimation to generate the specific grasp more accurately. Combining all together, the EconomicGrasp framework is proposed. Our extensive experiments show that EconomicGrasp surpasses the SOTA grasp method by about 3AP on average, and with extremely low resource cost, for about 1/4 training time cost, 1/8 memory cost and 1/30 storage cost. Our code is available at https://github.com/iSEE-Laboratory/EconomicGrasp.

An-Lan Wang, Kun-Yu Lin, Jia-Run Du et al. · tencent-ai

In this work, we focus on the task of procedure planning from instructional videos with text supervision, where a model aims to predict an action sequence to transform the initial visual state into the goal visual state. A critical challenge of this task is the large semantic gap between observed visual states and unobserved intermediate actions, which is ignored by previous works. Specifically, this semantic gap refers to that the contents in the observed visual states are semantically different from the elements of some action text labels in a procedure. To bridge this semantic gap, we propose a novel event-guided paradigm, which first infers events from the observed states and then plans out actions based on both the states and predicted events. Our inspiration comes from that planning a procedure from an instructional video is to complete a specific event and a specific event usually involves specific actions. Based on the proposed paradigm, we contribute an Event-guided Prompting-based Procedure Planning (E3P) model, which encodes event information into the sequential modeling process to support procedure planning. To further consider the strong action associations within each event, our E3P adopts a mask-and-predict approach for relation mining, incorporating a probabilistic masking scheme for regularization. Extensive experiments on three datasets demonstrate the effectiveness of our proposed model.

Ganzhao Yuan, Li Shen, Wei-Shi Zheng

The sparse generalized eigenvalue problem arises in a number of standard and modern statistical learning models, including sparse principal component analysis, sparse Fisher discriminant analysis, and sparse canonical correlation analysis. However, this problem is difficult to solve since it is NP-hard. In this paper, we consider a new decomposition method to tackle this problem. Specifically, we use random or/and swapping strategies to find a working set and perform global combinatorial search over the small subset of variables. We consider a bisection search method and a coordinate descent method for solving the quadratic fractional programming subproblem. In addition, we provide some theoretical analysis for the proposed method. Our experiments have shown that the proposed method significantly and consistently outperforms existing solutions in term of accuracy.

Junjin Xiao, Yandan Yang, Xinyuan Chang et al.

Vision-Language-Action (VLA) models trained via imitation learning suffer from significant performance degradation in data-scarce scenarios due to their reliance on large-scale demonstration datasets. Although reinforcement learning (RL)-based post-training has proven effective in addressing data scarcity, its application to VLA models is hindered by the non-resettable nature of real-world environments. This limitation is particularly critical in high-risk domains such as industrial automation, where interactions often induce state changes that are costly or infeasible to revert. Furthermore, existing VLA approaches lack a reliable mechanism for detecting task completion, leading to redundant actions that reduce overall task success rates. To address these challenges, we propose RehearseVLA:, an RL-based post-training framework that replaces physical interaction with a low-cost world model-based virtual simulator. RehearseVLA: consists of two key components: (1) a physically-consistent world simulator that generates temporally consistent future visual observations, and (2) a vision-language model (VLM)-guided instant reflector that provides continuous reward signals and predicts action termination. This simulated environment enables VLA models to safely explore and generalize beyond their initial imitation learning distribution. Our method achieves notable performance gains with as few as five expert demonstrations per task. Experiments on complex robotic manipulation tasks demonstrate that RehearseVLA: effectively overcomes the data inefficiency, safety constraints, and inefficient execution of conventional VLA models that rely on real-world interaction, offering a practical and scalable solution for post-training in resource-constrained settings. Our code is available at https://github.com/amap-cvlab/world-env.

Ganzhao Yuan, Li Shen, Wei-Shi Zheng

Sparse optimization is a central problem in machine learning and computer vision. However, this problem is inherently NP-hard and thus difficult to solve in general. Combinatorial search methods find the global optimal solution but are confined to small-sized problems, while coordinate descent methods are efficient but often suffer from poor local minima. This paper considers a new block decomposition algorithm that combines the effectiveness of combinatorial search methods and the efficiency of coordinate descent methods. Specifically, we consider a random strategy or/and a greedy strategy to select a subset of coordinates as the working set, and then perform a global combinatorial search over the working set based on the original objective function. We show that our method finds stronger stationary points than Amir Beck et al.'s coordinate-wise optimization method. In addition, we establish the convergence rate of our algorithm. Our experiments on solving sparse regularized and sparsity constrained least squares optimization problems demonstrate that our method achieves state-of-the-art performance in terms of accuracy. For example, our method generally outperforms the well-known greedy pursuit method.

Wentao Zhang, Yujun Huang, Tong Zhang et al.

Currently intelligent diagnosis systems lack the ability of continually learning to diagnose new diseases once deployed, under the condition of preserving old disease knowledge. In particular, updating an intelligent diagnosis system with training data of new diseases would cause catastrophic forgetting of old disease knowledge. To address the catastrophic forgetting issue, an Adapter-based Continual Learning framework called ACL is proposed to help effectively learn a set of new diseases at each round (or task) of continual learning, without changing the shared feature extractor. The learnable lightweight task-specific adapter(s) can be flexibly designed (e.g., two convolutional layers) and then added to the pretrained and fixed feature extractor. Together with a specially designed task-specific head which absorbs all previously learned old diseases as a single "out-of-distribution" category, task-specific adapter(s) can help the pretrained feature extractor more effectively extract discriminative features between diseases. In addition, a simple yet effective fine-tuning is applied to collaboratively fine-tune multiple task-specific heads such that outputs from different heads are comparable and consequently the appropriate classifier head can be more accurately selected during model inference. Extensive empirical evaluations on three image datasets demonstrate the superior performance of ACL in continual learning of new diseases. The source code is available at https://github.com/GiantJun/CL_Pytorch.

Renjie Lu, Jingke Meng, Wei-Shi Zheng

Vision and language navigation is a task that requires an agent to navigate according to a natural language instruction. Recent methods predict sub-goals on constructed topology map at each step to enable long-term action planning. However, they suffer from high computational cost when attempting to support such high-level predictions with GCN-like models. In this work, we propose an alternative method that facilitates navigation planning by considering the alignment between instructions and directed fidelity trajectories, which refers to a path from the initial node to the candidate locations on a directed graph without detours. This planning strategy leads to an efficient model while achieving strong performance. Specifically, we introduce a directed graph to illustrate the explored area of the environment, emphasizing directionality. Then, we firstly define the trajectory representation as a sequence of directed edge features, which are extracted from the panorama based on the corresponding orientation. Ultimately, we assess and compare the alignment between instruction and different trajectories during navigation to determine the next navigation target. Our method outperforms previous SOTA method BEVBert on RxR dataset and is comparable on R2R dataset while largely reducing the computational cost. Code is available: https://github.com/iSEE-Laboratory/VLN-PRET.

Jiawei Feng, Ancong Wu, Wei-Shi Zheng

Due to the modality gap between visible and infrared images with high visual ambiguity, learning \textbf{diverse} modality-shared semantic concepts for visible-infrared person re-identification (VI-ReID) remains a challenging problem. Body shape is one of the significant modality-shared cues for VI-ReID. To dig more diverse modality-shared cues, we expect that erasing body-shape-related semantic concepts in the learned features can force the ReID model to extract more and other modality-shared features for identification. To this end, we propose shape-erased feature learning paradigm that decorrelates modality-shared features in two orthogonal subspaces. Jointly learning shape-related feature in one subspace and shape-erased features in the orthogonal complement achieves a conditional mutual information maximization between shape-erased feature and identity discarding body shape information, thus enhancing the diversity of the learned representation explicitly. Extensive experiments on SYSU-MM01, RegDB, and HITSZ-VCM datasets demonstrate the effectiveness of our method.

Xin Zhang, Xiaohua Xie, Jianhuang Lai et al.

We are concerned with retrieving a query person from multiple videos captured by a non-overlapping camera network. Existing methods often rely on purely visual matching or consider temporal constraints but ignore the spatial information of the camera network. To address this issue, we propose a pedestrian retrieval framework based on cross-camera trajectory generation, which integrates both temporal and spatial information. To obtain pedestrian trajectories, we propose a novel cross-camera spatio-temporal model that integrates pedestrians' walking habits and the path layout between cameras to form a joint probability distribution. Such a spatio-temporal model among a camera network can be specified using sparsely sampled pedestrian data. Based on the spatio-temporal model, cross-camera trajectories can be extracted by the conditional random field model and further optimized by restricted non-negative matrix factorization. Finally, a trajectory re-ranking technique is proposed to improve the pedestrian retrieval results. To verify the effectiveness of our method, we construct the first cross-camera pedestrian trajectory dataset, the Person Trajectory Dataset, in real surveillance scenarios. Extensive experiments verify the effectiveness and robustness of the proposed method.

Yuan-Ming Li, Qize Yang, Nan Lei et al.

Recent advances in motion-aware large language models have shown remarkable promise for unifying motion understanding and generation tasks. However, these models typically treat understanding and generation separately, limiting the mutual benefits that could arise from interactive feedback between tasks. In this work, we reveal that motion assessment and refinement tasks act as crucial bridges to enable bidirectional knowledge flow between understanding and generation. Leveraging this insight, we propose Interleaved Reasoning for Motion Generation (IRMoGen), a novel paradigm that tightly couples motion generation with assessment and refinement through iterative text-motion dialogue. To realize this, we introduce IRG-MotionLLM, the first model that seamlessly interleaves motion generation, assessment, and refinement to improve generation performance. IRG-MotionLLM is developed progressively with a novel three-stage training scheme, initializing and subsequently enhancing native IRMoGen capabilities. To facilitate this development, we construct an automated data engine to synthesize interleaved reasoning annotations from existing text-motion datasets. Extensive experiments demonstrate that: (i) Assessment and refinement tasks significantly improve text-motion alignment; (ii) Interleaving motion generation, assessment, and refinement steps yields consistent performance gains across training stages; and (iii) IRG-MotionLLM clearly outperforms the baseline model and achieves advanced performance on standard text-to-motion generation benchmarks. Cross-evaluator testing further validates its effectiveness. Code & Data: https://github.com/HumanMLLM/IRG-MotionLLM/tree/main.

Hanjun Li, Xingjia Pan, Ke Yan et al.

Object detection under imperfect data receives great attention recently. Weakly supervised object detection (WSOD) suffers from severe localization issues due to the lack of instance-level annotation, while semi-supervised object detection (SSOD) remains challenging led by the inter-image discrepancy between labeled and unlabeled data. In this study, we propose the Single Instance annotated Object Detection (SIOD), requiring only one instance annotation for each existing category in an image. Degraded from inter-task (WSOD) or inter-image (SSOD) discrepancies to the intra-image discrepancy, SIOD provides more reliable and rich prior knowledge for mining the rest of unlabeled instances and trades off the annotation cost and performance. Under the SIOD setting, we propose a simple yet effective framework, termed Dual-Mining (DMiner), which consists of a Similarity-based Pseudo Label Generating module (SPLG) and a Pixel-level Group Contrastive Learning module (PGCL). SPLG firstly mines latent instances from feature representation space to alleviate the annotation missing problem. To avoid being misled by inaccurate pseudo labels, we propose PGCL to boost the tolerance to false pseudo labels. Extensive experiments on MS COCO verify the feasibility of the SIOD setting and the superiority of the proposed method, which obtains consistent and significant improvements compared to baseline methods and achieves comparable results with fully supervised object detection (FSOD) methods with only 40% instances annotated.

Yu-Ming Tang, Yi-Xing Peng, Wei-Shi Zheng

Deep neural network (DNN) suffers from catastrophic forgetting when learning incrementally, which greatly limits its applications. Although maintaining a handful of samples (called `exemplars`) of each task could alleviate forgetting to some extent, existing methods are still limited by the small number of exemplars since these exemplars are too few to carry enough task-specific knowledge, and therefore the forgetting remains. To overcome this problem, we propose to `imagine` diverse counterparts of given exemplars referring to the abundant semantic-irrelevant information from unlabeled data. Specifically, we develop a learnable feature generator to diversify exemplars by adaptively generating diverse counterparts of exemplars based on semantic information from exemplars and semantically-irrelevant information from unlabeled data. We introduce semantic contrastive learning to enforce the generated samples to be semantic consistent with exemplars and perform semanticdecoupling contrastive learning to encourage diversity of generated samples. The diverse generated samples could effectively prevent DNN from forgetting when learning new tasks. Our method does not bring any extra inference cost and outperforms state-of-the-art methods on two benchmarks CIFAR-100 and ImageNet-Subset by a clear margin.

Jiayu Jiao, Yu-Ming Tang, Kun-Yu Lin et al.

As a de facto solution, the vanilla Vision Transformers (ViTs) are encouraged to model long-range dependencies between arbitrary image patches while the global attended receptive field leads to quadratic computational cost. Another branch of Vision Transformers exploits local attention inspired by CNNs, which only models the interactions between patches in small neighborhoods. Although such a solution reduces the computational cost, it naturally suffers from small attended receptive fields, which may limit the performance. In this work, we explore effective Vision Transformers to pursue a preferable trade-off between the computational complexity and size of the attended receptive field. By analyzing the patch interaction of global attention in ViTs, we observe two key properties in the shallow layers, namely locality and sparsity, indicating the redundancy of global dependency modeling in shallow layers of ViTs. Accordingly, we propose Multi-Scale Dilated Attention (MSDA) to model local and sparse patch interaction within the sliding window. With a pyramid architecture, we construct a Multi-Scale Dilated Transformer (DilateFormer) by stacking MSDA blocks at low-level stages and global multi-head self-attention blocks at high-level stages. Our experiment results show that our DilateFormer achieves state-of-the-art performance on various vision tasks. On ImageNet-1K classification task, DilateFormer achieves comparable performance with 70% fewer FLOPs compared with existing state-of-the-art models. Our DilateFormer-Base achieves 85.6% top-1 accuracy on ImageNet-1K classification task, 53.5% box mAP/46.1% mask mAP on COCO object detection/instance segmentation task and 51.1% MS mIoU on ADE20K semantic segmentation task.

Ganzhao Yuan, Wei-Shi Zheng, Li Shen et al.

Composite function minimization captures a wide spectrum of applications in both computer vision and machine learning. It includes bound constrained optimization, $\ell_1$ norm regularized optimization, and $\ell_0$ norm regularized optimization as special cases. This paper proposes and analyzes a new Generalized Matrix Splitting Algorithm (GMSA) for minimizing composite functions. It can be viewed as a generalization of the classical Gauss-Seidel method and the Successive Over-Relaxation method for solving linear systems in the literature. Our algorithm is derived from a novel triangle operator mapping, which can be computed exactly using a new generalized Gaussian elimination procedure. We establish the global convergence, convergence rate, and iteration complexity of GMSA for convex problems. In addition, we also discuss several important extensions of GMSA. Finally, we validate the performance of our proposed method on three particular applications: nonnegative matrix factorization, $\ell_0$ norm regularized sparse coding, and $\ell_1$ norm regularized Dantzig selector problem. Extensive experiments show that our method achieves state-of-the-art performance in term of both efficiency and efficacy.

Jianxiong Tang, Jianhuang Lai, Xiaohua Xie et al.

Spiking neural networks are efficient computation models for low-power environments. Spike-based BP algorithms and ANN-to-SNN (ANN2SNN) conversions are successful techniques for SNN training. Nevertheless, the spike-base BP training is slow and requires large memory costs. Though ANN2NN provides a low-cost way to train SNNs, it requires many inference steps to mimic the well-trained ANN for good performance. In this paper, we propose a SNN-to-ANN (SNN2ANN) framework to train the SNN in a fast and memory-efficient way. The SNN2ANN consists of 2 components: a) a weight sharing architecture between ANN and SNN and b) spiking mapping units. Firstly, the architecture trains the weight-sharing parameters on the ANN branch, resulting in fast training and low memory costs for SNN. Secondly, the spiking mapping units ensure that the activation values of the ANN are the spiking features. As a result, the classification error of the SNN can be optimized by training the ANN branch. Besides, we design an adaptive threshold adjustment (ATA) algorithm to address the noisy spike problem. Experiment results show that our SNN2ANN-based models perform well on the benchmark datasets (CIFAR10, CIFAR100, and Tiny-ImageNet). Moreover, the SNN2ANN can achieve comparable accuracy under 0.625x time steps, 0.377x training time, 0.27x GPU memory costs, and 0.33x spike activities of the Spike-based BP model.

Kanghao Chen, Weixian Lei, Rong Zhang et al.

Imbalanced training data is a significant challenge for medical image classification. In this study, we propose a novel Progressive Class-Center Triplet (PCCT) framework to alleviate the class imbalance issue particularly for diagnosis of rare diseases, mainly by carefully designing the triplet sampling strategy and the triplet loss formation. Specifically, the PCCT framework includes two successive stages. In the first stage, PCCT trains the diagnosis system via a class-balanced triplet loss to coarsely separate distributions of different classes. In the second stage, the PCCT framework further improves the diagnosis system via a class-center involved triplet loss to cause a more compact distribution for each class. For the class-balanced triplet loss, triplets are sampled equally for each class at each training iteration, thus alleviating the imbalanced data issue. For the class-center involved triplet loss, the positive and negative samples in each triplet are replaced by their corresponding class centers, which enforces data representations of the same class closer to the class center. Furthermore, the class-center involved triplet loss is extended to the pair-wise ranking loss and the quadruplet loss, which demonstrates the generalization of the proposed framework. Extensive experiments support that the PCCT framework works effectively for medical image classification with imbalanced training images. On two skin image datasets and one chest X-ray dataset, the proposed approach respectively obtains the mean F1 score 86.2, 65.2, and 90.66 over all classes and 81.4, 63.87, and 81.92 for rare classes, achieving state-of-the-art performance and outperforming the widely used methods for the class imbalance issue.

Jia-Xin Zhuang, Jiabin Cai, Jianguo Zhang et al.

Automated medical image classification is the key component in intelligent diagnosis systems. However, most medical image datasets contain plenty of samples of common diseases and just a handful of rare ones, leading to major class imbalances. Currently, it is an open problem in intelligent diagnosis to effectively learn from imbalanced training data. In this paper, we propose a simple yet effective framework, named \textbf{C}lass \textbf{A}ttention to \textbf{RE}gions of the lesion (CARE), to handle data imbalance issues by embedding attention into the training process of \textbf{C}onvolutional \textbf{N}eural \textbf{N}etworks (CNNs). The proposed attention module helps CNNs attend to lesion regions of rare diseases, therefore helping CNNs to learn their characteristics more effectively. In addition, this attention module works only during the training phase and does not change the architecture of the original network, so it can be directly combined with any existing CNN architecture. The CARE framework needs bounding boxes to represent the lesion regions of rare diseases. To alleviate the need for manual annotation, we further developed variants of CARE by leveraging the traditional saliency methods or a pretrained segmentation model for bounding box generation. Results show that the CARE variants with automated bounding box generation are comparable to the original CARE framework with \textit{manual} bounding box annotations. A series of experiments on an imbalanced skin image dataset and a pneumonia dataset indicates that our method can effectively help the network focus on the lesion regions of rare diseases and remarkably improves the classification performance of rare diseases.

Yipeng Gao, Lingxiao Yang, Yunmu Huang et al.

Under the domain shift, cross-domain few-shot object detection aims to adapt object detectors in the target domain with a few annotated target data. There exists two significant challenges: (1) Highly insufficient target domain data; (2) Potential over-adaptation and misleading caused by inappropriately amplified target samples without any restriction. To address these challenges, we propose an adaptive method consisting of two parts. First, we propose an adaptive optimization strategy to select augmented data similar to target samples rather than blindly increasing the amount. Specifically, we filter the augmented candidates which significantly deviate from the target feature distribution in the very beginning. Second, to further relieve the data limitation, we propose the multi-level domain-aware data augmentation to increase the diversity and rationality of augmented data, which exploits the cross-image foreground-background mixture. Experiments show that the proposed method achieves state-of-the-art performance on multiple benchmarks.

An-Lan Wang, Bin Shan, Wei Shi et al.

This work presents ParGo, a novel Partial-Global projector designed to connect the vision and language modalities for Multimodal Large Language Models (MLLMs). Unlike previous works that rely on global attention-based projectors, our ParGo bridges the representation gap between the separately pre-trained vision encoders and the LLMs by integrating global and partial views, which alleviates the overemphasis on prominent regions. To facilitate the effective training of ParGo, we collect a large-scale detail-captioned image-text dataset named ParGoCap-1M-PT, consisting of 1 million images paired with high-quality captions. Extensive experiments on several MLLM benchmarks demonstrate the effectiveness of our ParGo, highlighting its superiority in aligning vision and language modalities. Compared to conventional Q-Former projector, our ParGo achieves an improvement of 259.96 in MME benchmark. Furthermore, our experiments reveal that ParGo significantly outperforms other projectors, particularly in tasks that emphasize detail perception ability.

Tianming Liang, Qirui Du, Jian-Fang Hu et al.

Segmentation based on language has been a popular topic in computer vision. While recent advances in multimodal large language models (MLLMs) have endowed segmentation systems with reasoning capabilities, these efforts remain confined by the frozen internal knowledge of MLLMs, which limits their potential for real-world scenarios that involve up-to-date information or domain-specific concepts. In this work, we propose \textbf{Seg-ReSearch}, a novel segmentation paradigm that overcomes the knowledge bottleneck of existing approaches. By enabling interleaved reasoning and external search, Seg-ReSearch empowers segmentation systems to handle dynamic, open-world queries that extend beyond the frozen knowledge of MLLMs. To effectively train this capability, we introduce a hierarchical reward design that harmonizes initial guidance with progressive incentives, mitigating the dilemma between sparse outcome signals and rigid step-wise supervision. For evaluation, we construct OK-VOS, a challenging benchmark that explicitly requires outside knowledge for video object segmentation. Experiments on OK-VOS and two existing reasoning segmentation benchmarks demonstrate that our Seg-ReSearch improves state-of-the-art approaches by a substantial margin. Code and data will be released at https://github.com/iSEE-Laboratory/Seg-ReSearch.

Nan Lei, Yuan-Ming Li, Ling-An Zeng et al.

Despite substantial progress in text-driven 3D human motion synthesis, generating realistic multi-person interaction sequences remains challenging. Notably, body inter-penetration is a pervasive issue from both data acquisition to the generated results, which significantly undermines the realism and usability. Previous generative models either ignored this issue or introduced computationally expensive mesh-level loss functions to alleviate inter-body collisions. In this paper, we propose a general-purpose and computationally efficient optimization strategy named PhysiGen to explicitly integrate collision-aware physical constraints for human-human interaction generation. Specifically, we simplify the high-resolution human body mesh into geometric primitives to greatly reduce the cost of inter-person collision detection. Moreover, we identify the collision regions as the guidance of the optimization directions. PhysiGen is plug-and-play and can be readily integrated into existing human interaction generation models. Extensive cross-dataset and cross-model experiments show that our method can effectively reduce interpenetration and significantly improve visual coherence and physical plausibility compared to the state-of-the-art methods.

Sheng Xu, Guiliang Liu, Tarak Kharrat et al.

Success in association football relies on both individual skill and coordinated tactics. While recent advancements in spatio-temporal data and deep learning have enabled predictive analyses like trajectory forecasting, the development of tactical design remains limited. Bridging this gap is essential, as prediction reveals what is likely to occur, whereas tactic generation determines what should occur to achieve strategic objectives. In this work, we present TacticGen, a generative model for adaptable and scalable tactic generation. TacticGen formulates tactics as sequences of multi-agent movements and interactions conditioned on the game context. It employs a multi-agent diffusion transformer with agent-wise self-attention and context-aware cross-attention to capture cooperative and competitive dynamics among players and the ball. Trained with over 3.3 million events and 100 million tracking frames from top-tier leagues, TacticGen achieves state-of-the-art precision in predicting player trajectories. Building on it, TacticGen enables adaptable tactic generation tailored to diverse inference-time objectives through classifier guidance mechanism, specified via rules, natural language, or neural models. Its modeling performance is also inherently scalable. A case study with football experts confirms that TacticGen generates realistic, strategically valuable tactics, demonstrating its practical utility for tactical planning in professional football. The project page is available at: https://shengxu.net/TacticGen/.

Haoxuan Chen, Tianming Liang, Wei-Shi Zheng et al.

Reinforcement learning with verifiers (RLVR) has become a central paradigm for improving LLM reasoning, yet popular group-based optimization algorithms like GRPO often suffer from exploration collapse, where the models prematurely converge on a narrow set of high-scoring patterns, lacking the ability to explore new solutions. Recent efforts attempt to alleviate this by adding entropy regularization or diversity bonus. However, these approaches do not change the \textit{winner-takes-all} nature, where rollouts still compete for individual advantage rather than cooperating for maximizing global diversity. In this work, we propose Group Cooperative Policy Optimization (GCPO), which shifts the training paradigm from rollout competition to team cooperation. Specifically, GCPO replaces independent rollout scoring with team-level credit assignment: a rollout is rewarded by how much it contributes to the team's valid solution coverage, rather than its individual accuracy. This coverage is described as a determinant volume over reward-weighted semantic embeddings, where only correct and non-redundant rollouts contribute to this volume. During advantage estimation, GCPO redistributes the collective team reward to each single rollout according to its average marginal contribution to the team. This cooperative training paradigm routes optimization toward non-redundant correct reasoning paths. Experiments across multiple reasoning benchmarks demonstrate that GCPO significantly improves both reasoning accuracy and solution diversity over existing approaches. Code will be released at $\href{https://github.com/bradybuddiemarch/gcpo}{this}$.

Xuankai Zhang, Junjin Xiao, Shangwei Huang et al.

We present an approach for high-quality dynamic Gaussian Splatting from monocular videos. To this end, we in this work go one step further beyond previous methods to explicitly model continuous position and orientation deformation of dynamic Gaussians, using an SE(3) B-spline motion bases with a compact set of control points. To improve computational efficiency while enhancing the ability to model complex motions, an adaptive control mechanism is devised to dynamically adjust the number of motion bases and control points. Besides, we develop a soft segment reconstruction strategy to mitigate long-interval motion interference, and employ a multi-view diffusion model to provide multi-view cues for avoiding overfitting to training views. Extensive experiments demonstrate that our method outperforms state-of-the-art methods in novel view synthesis. Our code is available at https://github.com/hhhddddddd/se3bsplinegs.

Kun-Yu Lin, Jia-Run Du, Yipeng Gao et al.

Video domain generalization aims to learn generalizable video classification models for unseen target domains by training in a source domain. A critical challenge of video domain generalization is to defend against the heavy reliance on domain-specific cues extracted from the source domain when recognizing target videos. To this end, we propose to perceive diverse spatial-temporal cues in videos, aiming to discover potential domain-invariant cues in addition to domain-specific cues. We contribute a novel model named Spatial-Temporal Diversification Network (STDN), which improves the diversity from both space and time dimensions of video data. First, our STDN proposes to discover various types of spatial cues within individual frames by spatial grouping. Then, our STDN proposes to explicitly model spatial-temporal dependencies between video contents at multiple space-time scales by spatial-temporal relation modeling. Extensive experiments on three benchmarks of different types demonstrate the effectiveness and versatility of our approach.

Dian Zheng, Xiao-Ming Wu, Zuhao Liu et al.

Stereo matching is a significant part in many computer vision tasks and driving-based applications. Recently cost volume-based methods have achieved great success benefiting from the rich geometry information in paired images. However, the redundancy of cost volume also interferes with the model training and limits the performance. To construct a more precise cost volume, we pioneeringly apply the diffusion model to stereo matching. Our method, termed DiffuVolume, considers the diffusion model as a cost volume filter, which will recurrently remove the redundant information from the cost volume. Two main designs make our method not trivial. Firstly, to make the diffusion model more adaptive to stereo matching, we eschew the traditional manner of directly adding noise into the image but embed the diffusion model into a task-specific module. In this way, we outperform the traditional diffusion stereo matching method by 22% EPE improvement and 240 times inference acceleration. Secondly, DiffuVolume can be easily embedded into any volume-based stereo matching network with boost performance but slight parameters rise (only 2%). By adding the DiffuVolume into well-performed methods, we outperform all the published methods on Scene Flow, KITTI2012, KITTI2015 benchmarks and zero-shot generalization setting. It is worth mentioning that the proposed model ranks 1st on KITTI 2012 leader board, 2nd on KITTI 2015 leader board since 15, July 2023.

Zihang Lin, Chaolei Tan, Jian-Fang Hu et al.

In this technical report, we introduce our solution to human-centric spatio-temporal video grounding task. We propose a concise and effective framework named STVGFormer, which models spatiotemporal visual-linguistic dependencies with a static branch and a dynamic branch. The static branch performs cross-modal understanding in a single frame and learns to localize the target object spatially according to intra-frame visual cues like object appearances. The dynamic branch performs cross-modal understanding across multiple frames. It learns to predict the starting and ending time of the target moment according to dynamic visual cues like motions. Both the static and dynamic branches are designed as cross-modal transformers. We further design a novel static-dynamic interaction block to enable the static and dynamic branches to transfer useful and complementary information from each other, which is shown to be effective to improve the prediction on hard cases. Our proposed method achieved 39.6% vIoU and won the first place in the HC-STVG track of the 4th Person in Context Challenge.

Jiahao Niu, Rongjia Zheng, Wenju Xu et al.

We present SGS-Intrinsic, an indoor inverse rendering framework that works well for sparse-view images. Unlike existing 3D Gaussian Splatting (3DGS) based methods that focus on object-centric reconstruction and fail to work under sparse view settings, our method allows to achieve high-quality geometry reconstruction and accurate disentanglement of material and illumination. The core idea is to construct a dense and geometry-consistent Gaussian semantic field guided by semantic and geometric priors, providing a reliable foundation for subsequent inverse rendering. Building upon this, we perform material-illumination disentanglement by combining a hybrid illumination model and material prior to effectively capture illumination-material interactions. To mitigate the impact of cast shadows and enhance the robustness of material recovery, we introduce illumination-invariant material constraint together with a deshadowing model. Extensive experiments on benchmark datasets show that our method consistently improves both reconstruction fidelity and inverse rendering quality over existing 3DGS-based inverse rendering approaches. Our code is available at https://github.com/GrumpySloths/SGS_Intrinsic.github.io.

Peipei Zhu, Xiao Wang, Yong Luo et al.

The goal of unpaired image captioning (UIC) is to describe images without using image-caption pairs in the training phase. Although challenging, we except the task can be accomplished by leveraging a training set of images aligned with visual concepts. Most existing studies use off-the-shelf algorithms to obtain the visual concepts because the Bounding Box (BBox) labels or relationship-triplet labels used for the training are expensive to acquire. In order to resolve the problem in expensive annotations, we propose a novel approach to achieve cost-effective UIC. Specifically, we adopt image-level labels for the optimization of the UIC model in a weakly-supervised manner. For each image, we assume that only the image-level labels are available without specific locations and numbers. The image-level labels are utilized to train a weakly-supervised object recognition model to extract object information (e.g., instance) in an image, and the extracted instances are adopted to infer the relationships among different objects based on an enhanced graph neural network (GNN). The proposed approach achieves comparable or even better performance compared with previous methods without the expensive cost of annotations. Furthermore, we design an unrecognized object (UnO) loss combined with a visual concept reward to improve the alignment of the inferred object and relationship information with the images. It can effectively alleviate the issue encountered by existing UIC models about generating sentences with nonexistent objects. To the best of our knowledge, this is the first attempt to solve the problem of Weakly-Supervised visual concept recognition for UIC (WS-UIC) based only on image-level labels. Extensive experiments have been carried out to demonstrate that the proposed WS-UIC model achieves inspiring results on the COCO dataset while significantly reducing the cost of labeling.

Rongjia Zheng, Shangwei Huang, Lei Zhu et al.

We present a generative method for texture filtering, which exhibits surprisingly good performance and generalizability. Our core idea is to empower texture filtering by taking full advantage of the strong learned image prior of pre-trained generative models. To this end, we propose to fine-tune a pre-trained generative model via a two-stage strategy. Specifically, we first conduct supervised fine-tuning on a very small set of paired images, and then perform reinforcement fine-tuning on a large-scale unlabeled dataset under the guidance of a reward function that quantifies the quality of texture removal and structure preservation. Extensive experiments show that our method clearly outperforms previous methods, and is effective to deal with previously challenging cases. Our code is available at https://github.com/OnlyZZZZ/Generative_Texture_Filtering.

Chaolei Tan, Zihang Lin, Junfu Pu et al.

Video grounding is a fundamental problem in multimodal content understanding, aiming to localize specific natural language queries in an untrimmed video. However, current video grounding datasets merely focus on simple events and are either limited to shorter videos or brief sentences, which hinders the model from evolving toward stronger multimodal understanding capabilities. To address these limitations, we present a large-scale video grounding dataset named SynopGround, in which more than 2800 hours of videos are sourced from popular TV dramas and are paired with accurately localized human-written synopses. Each paragraph in the synopsis serves as a language query and is manually annotated with precise temporal boundaries in the long video. These paragraph queries are tightly correlated to each other and contain a wealth of abstract expressions summarizing video storylines and specific descriptions portraying event details, which enables the model to learn multimodal perception on more intricate concepts over longer context dependencies. Based on the dataset, we further introduce a more complex setting of video grounding dubbed Multi-Paragraph Video Grounding (MPVG), which takes as input multiple paragraphs and a long video for grounding each paragraph query to its temporal interval. In addition, we propose a novel Local-Global Multimodal Reasoner (LGMR) to explicitly model the local-global structures of long-term multimodal inputs for MPVG. Our method provides an effective baseline solution to the multi-paragraph video grounding problem. Extensive experiments verify the proposed model's effectiveness as well as its superiority in long-term multi-paragraph video grounding over prior state-of-the-arts. Dataset and code are publicly available. Project page: https://synopground.github.io/.

Yuliang Wu, Yanhan Lin, WengKit Lao et al.

To meet the demands of increasingly diverse dexterous hand hardware, it is crucial to develop a policy that enables zero-shot cross-embodiment grasping without redundant re-learning. Cross-embodiment alignment is challenging due to heterogeneous hand kinematics and physical constraints. Existing approaches typically predict intermediate motion targets and retarget them to each embodiment, which may introduce errors and violate embodiment-specific limits, hindering transfer across diverse hands. To overcome these limitations, we propose DexGrasp-Zero, a policy that learns universal grasping skills from diverse embodiments, enabling zero-shot transfer to unseen hands. We first introduce a morphology-aligned graph representation that maps each hand's kinematic keypoints to anatomically grounded nodes and equips each node with tri-axial orthogonal motion primitives, enabling structural and semantic alignment across different morphologies. Relying on this graph-based representation, we design a Morphology-Aligned Graph Convolutional Network (MAGCN) to encode the graph for policy learning. MAGCN incorporates a Physical Property Injection mechanism that fuses hand-specific physical constraints into the graph features, enabling adaptive compensation for varying link lengths and actuation limits for precise and stable grasping. Our extensive simulation evaluations on the YCB dataset demonstrate that our policy, jointly trained on four heterogeneous hands (Allegro, Shadow, Schunk, Ability), achieves an 85% zero-shot success rate on unseen hardware (LEAP, Inspire), outperforming the state-of-the-art method by 59.5%. Real-world experiments further evaluate our policy on three robot platforms (LEAP, Inspire, Revo2), achieving an 82% average success rate on unseen objects.

Kun-Yu Lin, Jiaming Zhou, Wei-Shi Zheng

We study the domain adaptation task for action recognition, namely domain adaptive action recognition, which aims to effectively transfer action recognition power from a label-sufficient source domain to a label-free target domain. Since actions are performed by humans, it is crucial to exploit human cues in videos when recognizing actions across domains. However, existing methods are prone to losing human cues but prefer to exploit the correlation between non-human contexts and associated actions for recognition, and the contexts of interest agnostic to actions would reduce recognition performance in the target domain. To overcome this problem, we focus on uncovering human-centric action cues for domain adaptive action recognition, and our conception is to investigate two aspects of human-centric action cues, namely human cues and human-context interaction cues. Accordingly, our proposed Human-Centric Transformer (HCTransformer) develops a decoupled human-centric learning paradigm to explicitly concentrate on human-centric action cues in domain-variant video feature learning. Our HCTransformer first conducts human-aware temporal modeling by a human encoder, aiming to avoid a loss of human cues during domain-invariant video feature learning. Then, by a Transformer-like architecture, HCTransformer exploits domain-invariant and action-correlated contexts by a context encoder, and further models domain-invariant interaction between humans and action-correlated contexts. We conduct extensive experiments on three benchmarks, namely UCF-HMDB, Kinetics-NecDrone and EPIC-Kitchens-UDA, and the state-of-the-art performance demonstrates the effectiveness of our proposed HCTransformer.

Jiangxin Sun, Feng Xue, Teng Long et al.

With advances in imitation learning (IL) and large-scale driving datasets, end-to-end autonomous driving (E2E-AD) has made great progress recently. Currently, IL-based methods have become a mainstream paradigm: models rely on standard driving behaviors given by experts, and learn to minimize the discrepancy between their actions and expert actions. However, this objective of "only driving like the expert" suffers from limited generalization: when encountering rare or unseen long-tail scenarios outside the distribution of expert demonstrations, models tend to produce unsafe decisions in the absence of prior experience. This raises a fundamental question: Can an E2E-AD system make reliable decisions without any expert action supervision? Motivated by this, we propose a unified framework named Risk-aware World Model Predictive Control (RaWMPC) to address this generalization dilemma through robust control, without reliance on expert demonstrations. Practically, RaWMPC leverages a world model to predict the consequences of multiple candidate actions and selects low-risk actions through explicit risk evaluation. To endow the world model with the ability to predict the outcomes of risky driving behaviors, we design a risk-aware interaction strategy that systematically exposes the world model to hazardous behaviors, making catastrophic outcomes predictable and thus avoidable. Furthermore, to generate low-risk candidate actions at test time, we introduce a self-evaluation distillation method to distill riskavoidance capabilities from the well-trained world model into a generative action proposal network without any expert demonstration. Extensive experiments show that RaWMPC outperforms state-of-the-art methods in both in-distribution and out-of-distribution scenarios, while providing superior decision interpretability.

Chenghua Zeng, Huijuan Lu, Kanghao Chen et al.

Data imbalance between common and rare diseases during model training often causes intelligent diagnosis systems to have biased predictions towards common diseases. The state-of-the-art approaches apply a two-stage learning framework to alleviate the class-imbalance issue, where the first stage focuses on training of a general feature extractor and the second stage focuses on fine-tuning the classifier head for class rebalancing. However, existing two-stage approaches do not consider the fine-grained property between different diseases, often causing the first stage less effective for medical image classification than for natural image classification tasks. In this study, we propose embedding metric learning into the first stage of the two-stage framework specially to help the feature extractor learn to extract more discriminative feature representations. Extensive experiments mainly on three medical image datasets show that the proposed approach consistently outperforms existing onestage and two-stage approaches, suggesting that metric learning can be used as an effective plug-in component in the two-stage framework for fine-grained class-imbalanced image classification tasks.