Zhi Han

Nan Yang, Xin Luan, Huidi Jia et al.

Three problems exist in sequential facial image editing: incontinuous editing, inconsistent editing, and irreversible editing. Incontinuous editing is that the current editing can not retain the previously edited attributes. Inconsistent editing is that swapping the attribute editing orders can not yield the same results. Irreversible editing means that operating on a facial image is irreversible, especially in sequential facial image editing. In this work, we put forward three concepts and corresponding definitions: editing continuity, consistency, and reversibility. Then, we propose a novel model to achieve the goal of editing continuity, consistency, and reversibility. A sufficient criterion is defined to determine whether a model is continuous, consistent, and reversible. Extensive qualitative and quantitative experimental results validate our proposed model and show that a continuous, consistent and reversible editing model has a more flexible editing function while preserving facial identity. Furthermore, we think that our proposed definitions and model will have wide and promising applications in multimedia processing. Code and data are available at https://github.com/mickoluan/CCR.

Zihang Zhao, Zhenghao Qi, Yuyang Li et al. · pku

Adept manipulation of articulated objects is essential for robots to operate successfully in human environments. Such manipulation requires both effectiveness--reliable operation despite uncertain object structures--and efficiency--swift execution with minimal redundant steps and smooth actions. Existing approaches struggle to achieve both objectives simultaneously: methods relying on predefined kinematic models lack effectiveness when encountering structural variations, while tactile-informed approaches achieve robust manipulation without kinematic priors but compromise efficiency through reactive, step-by-step exploration-compensation cycles. This paper introduces TacMan-Turbo, a novel proactive tactile control framework for articulated object manipulation that mitigates this fundamental trade-off. Unlike previous approaches that treat tactile contact deviations merely as error signals requiring compensation, our method interprets these deviations as rich sources of local kinematic information. This new perspective enables our controller to predict optimal future interactions and make proactive adjustments, significantly enhancing manipulation efficiency. In comprehensive evaluations across 200 diverse simulated articulated objects and real-world experiments, our approach maintains a 100% success rate while significantly outperforming the previous tactile-informed method in time efficiency, action efficiency, and trajectory smoothness (all p-values < 0.0001). These results demonstrate that the long-standing trade-off between effectiveness and efficiency in articulated object manipulation can be successfully resolved without relying on prior kinematic knowledge.

Leiyao Cui, Zihang Zhao, Sirui Xie et al. · pku

Autonomous manipulation of articulated objects remains a fundamental challenge for robots in human environments. Vision-based methods can infer hidden kinematics but can yield imprecise estimates on unfamiliar objects. Tactile approaches achieve robust control through contact feedback but require accurate initialization. This suggests a natural synergy: vision for global guidance, touch for local precision. Yet no framework systematically exploits this complementarity for generalized articulated manipulation. Here we present Vi-TacMan, which uses vision to propose grasps and coarse directions that seed a tactile controller for precise execution. By incorporating surface normals as geometric priors and modeling directions via von Mises-Fisher distributions, our approach achieves significant gains over baselines (all p<0.0001). Critically, manipulation succeeds without explicit kinematic models -- the tactile controller refines coarse visual estimates through real-time contact regulation. Tests on more than 50,000 simulated and diverse real-world objects confirm robust cross-category generalization. This work establishes that coarse visual cues suffice for reliable manipulation when coupled with tactile feedback, offering a scalable paradigm for autonomous systems in unstructured environments.

Haoyang Luo, Zihang Zhao, Leiyao Cui et al. · pku

Human dexterity relies on rapid, sub-second motor adjustments, yet capturing these high-frequency dynamics remains an enduring challenge in biomechanics and robotics. Existing motion capture paradigms are compromised by a trade-off between temporal resolution and visual occlusion, failing to record the fine-grained hand motion of fast, contact-rich manipulation. Here we introduce T-800, a high-bandwidth data glove system that achieves synchronized, full-hand motion tracking at 800 Hz. By integrating a novel broadcast-based synchronization mechanism with a mechanical stress isolation architecture, our system maintains sub-frame temporal alignment across 18 distributed inertial measurement units (IMUs) during extended, vigorous movements. We demonstrate that T-800 recovers fine-grained manipulation details previously lost to temporal undersampling. Our analysis reveals that human dexterity exhibits significantly high-frequency motion energy (>100 Hz) that was fundamentally inaccessible due to the Nyquist sampling limit imposed by previous hardware constraints. To validate the system's utility for robotic manipulation, we implement a kinematic retargeting algorithm that maps T-800's high-fidelity human gestures onto dexterous robotic hand models. This demonstrates that the high-frequency motion data can be accurately translated while respecting the kinematic constraints of robotic hands, providing the rich behavioral data necessary for training robust control policies in the future.

Dongwei Wang, Zhi Han, Yanmei Wang et al.

Reviewing plays an important role when learning knowledge. The knowledge acquisition at a certain time point may be strongly inspired with the help of previous experience. Thus the knowledge growing procedure should show strong relationship along the temporal dimension. In our research, we find that during the network training, the evolution of feature map follows temporal sequence property. A proper temporal supervision may further improve the network training performance. Inspired by this observation, we propose Temporal Supervised Knowledge Distillation (TSKD). Specifically, we extract the spatiotemporal features in the different training phases of student by convolutional Long Short-term memory network (Conv-LSTM). Then, we train the student net through a dynamic target, rather than static teacher network features. This process realizes the refinement of old knowledge in student network, and utilizes it to assist current learning. Extensive experiments verify the effectiveness and advantages of our method over existing knowledge distillation methods, including various network architectures and different tasks (image classification and object detection) .

Weihong Ren, Denglu Wu, Hui Cao et al.

The recent trend in 2D multiple object tracking (MOT) is jointly solving detection and tracking, where object detection and appearance feature (or motion) are learned simultaneously. Despite competitive performance, in crowded scenes, joint detection and tracking usually fail to find accurate object associations due to missed or false detections. In this paper, we jointly model counting, detection and re-identification in an end-to-end framework, named CountingMOT, tailored for crowded scenes. By imposing mutual object-count constraints between detection and counting, the CountingMOT tries to find a balance between object detection and crowd density map estimation, which can help it to recover missed detections or reject false detections. Our approach is an attempt to bridge the gap of object detection, counting, and re-Identification. This is in contrast to prior MOT methods that either ignore the crowd density and thus are prone to failure in crowded scenes,or depend on local correlations to build a graphical relationship for matching targets. The proposed MOT tracker can perform online and real-time tracking, and achieves the state-of-the-art results on public benchmarks MOT16 (MOTA of 79.7), MOT17 (MOTA of 81.3%) and MOT20 (MOTA of 78.9%).

Zhi Han, Baichen Liu, Shao-Bo Lin et al.

This paper studies the performance of deep convolutional neural networks (DCNNs) with zero-padding in feature extraction and learning. After verifying the roles of zero-padding in enabling translation-equivalence, and pooling in its translation-invariance driven nature, we show that with similar number of free parameters, any deep fully connected networks (DFCNs) can be represented by DCNNs with zero-padding. This demonstrates that DCNNs with zero-padding is essentially better than DFCNs in feature extraction. Consequently, we derive universal consistency of DCNNs with zero-padding and show its translation-invariance in the learning process. All our theoretical results are verified by numerical experiments including both toy simulations and real-data running.

ZHiyu Liu, Haobo Geng, Xudong Wang et al.

We study the problem of recovering a low-tubal-rank tensor $\mathcal{X}\_\star\in \mathbb{R}^{n \times n \times k}$ from noisy linear measurements under the t-product framework. A widely adopted strategy involves factorizing the optimization variable as $\mathcal{U} * \mathcal{U}^\top$, where $\mathcal{U} \in \mathbb{R}^{n \times R \times k}$, followed by applying factorized gradient descent (FGD) to solve the resulting optimization problem. Since the tubal-rank $r$ of the underlying tensor $\mathcal{X}_\star$ is typically unknown, this method often assumes $r < R \le n$, a regime known as over-parameterization. However, when the measurements are corrupted by some dense noise (e.g., Gaussian noise), FGD with the commonly used spectral initialization yields a recovery error that grows linearly with the over-estimated tubal-rank $R$. To address this issue, we show that using a small initialization enables FGD to achieve a nearly minimax optimal recovery error, even when the tubal-rank $R$ is significantly overestimated. Using a four-stage analytic framework, we analyze this phenomenon and establish the sharpest known error bound to date, which is independent of the overestimated tubal-rank $R$. Furthermore, we provide a theoretical guarantee showing that an easy-to-use early stopping strategy can achieve the best known result in practice. All these theoretical findings are validated through a series of simulations and real-data experiments.

Zhiyu Liu, Zhi Han, Yandong Tang et al.

The problem of low-tubal-rank tensor estimation is a fundamental task with wide applications across high-dimensional signal processing, machine learning, and image science. Traditional approaches tackle such a problem by performing tensor singular value decomposition, which is computationally expensive and becomes infeasible for large-scale tensors. Recent approaches address this issue by factorizing the tensor into two smaller factor tensors and solving the resulting problem using gradient descent. However, this kind of approach requires an accurate estimate of the tensor rank, and when the rank is overestimated, the convergence of gradient descent and its variants slows down significantly or even diverges. To address this problem, we propose an Alternating Preconditioned Gradient Descent (APGD) algorithm, which accelerates convergence in the over-parameterized setting by adding a preconditioning term to the original gradient and updating these two factors alternately. Based on certain geometric assumptions on the objective function, we establish linear convergence guarantees for more general low-tubal-rank tensor estimation problems. Then we further analyze the specific cases of low-tubal-rank tensor factorization and low-tubal-rank tensor recovery. Our theoretical results show that APGD achieves linear convergence even under over-parameterization, and the convergence rate is independent of the tensor condition number. Extensive simulations on synthetic data are carried out to validate our theoretical assertions.

Xiaohang Nie, Zihan Guo, Zicai Cui et al.

As large language models (LLM)-driven agents transition from isolated task solvers to persistent digital entities, the emergence of the Agentic Web, an ecosystem where heterogeneous agents autonomously interact and co-evolve, marks a pivotal shift toward Artificial General Intelligence (AGI). However, LLM-based multi-agent systems (LaMAS) are hindered by open-world issues such as scaling friction, coordination breakdown, and value dissipation. To address these challenges, we introduce Holos, a web-scale LaMAS architected for long-term ecological persistence. Holos adopts a five-layer architecture, with core modules primarily featuring the Nuwa engine for high-efficiency agent generation and hosting, a market-driven Orchestrator for resilient coordination, and an endogenous value cycle to achieve incentive compatibility. By bridging the gap between micro-level collaboration and macro-scale emergence, Holos hopes to lay the foundation for the next generation of the self-organizing and continuously evolving Agentic Web. We have publicly released Holos (accessible at https://holosai.io), providing a resource for the community and a testbed for future research in large-scale agentic ecosystems.

Xudong Wang, Gan Li, Zhiyu Liu et al.

Deploying vision-and-language navigation (VLN) agents requires adaptation across diverse scenes and environments, but fine-tuning on a specific scenario often causes catastrophic forgetting in others, which severely limits flexible long-term deployment. We formalize this challenge as the all-day multi-scenes lifelong VLN (AML-VLN) problem. Existing parameter-efficient adapters (e.g., LoRA and its variants) are limited by their two-dimensional matrix form, which fails to capture the multi-hierarchical navigation knowledge spanning multiple scenes and environments. To address this, we propose Tucker Adaptation (TuKA), which represents the multi-hierarchical navigation knowledge as a high-order tensor and leverages Tucker decomposition to decouple the knowledge into shared subspaces and scenario-specific experts. We further introduce a decoupled knowledge incremental learning strategy to consolidate shared subspaces while constraining specific experts for decoupled lifelong learning. Building on TuKA, we also develop a VLN agent named AlldayWalker, which continually learns across multiple navigation scenarios, achieving all-day multi-scenes navigation. Extensive experiments show that AlldayWalker consistently outperforms state-of-the-art baselines.

Jiahua Dong, Xudong Wang, Wenqi Liang et al.

Customized text-to-video generation (CTVG) has recently witnessed great progress in generating tailored videos from user-specific text. However, most CTVG methods assume that personalized concepts remain static and do not expand incrementally over time. Additionally, they struggle with forgetting and concept neglect when continuously learning new concepts, including subjects and motions. To resolve the above challenges, we develop a novel Continual Customized Video Diffusion (CCVD) model, which can continuously learn new concepts to generate videos across various text-to-video generation tasks by tackling forgetting and concept neglect. To address catastrophic forgetting, we introduce a concept-specific attribute retention module and a task-aware concept aggregation strategy. They can capture the unique characteristics and identities of old concepts during training, while combining all subject and motion adapters of old concepts based on their relevance during testing. Besides, to tackle concept neglect, we develop a controllable conditional synthesis to enhance regional features and align video contexts with user conditions, by incorporating layer-specific region attention-guided noise estimation. Extensive experimental comparisons demonstrate that our CCVD outperforms existing CTVG baselines on both the DreamVideo and Wan 2.1 backbones. The code is available at https://github.com/JiahuaDong/CCVD.

Zebin Han, Xudong Wang, Baichen Liu et al.

Sequential-Horizon Vision-and-Language Navigation (SH-VLN) presents a challenging scenario where agents should sequentially execute multi-task navigation guided by complex, long-horizon language instructions. Current vision-and-language navigation models exhibit significant performance degradation with such multi-task instructions, as information overload impairs the agent's ability to attend to observationally relevant details. To address this problem, we propose SeqWalker, a navigation model built on a hierarchical planning framework. Our SeqWalker features: i) A High-Level Planner that dynamically selects global instructions into contextually relevant sub-instructions based on the agent's current visual observations, thus reducing cognitive load; ii) A Low-Level Planner incorporating an Exploration-Verification strategy that leverages the inherent logical structure of instructions for trajectory error correction. To evaluate SH-VLN performance, we also extend the IVLN dataset and establish a new benchmark. Extensive experiments are performed to demonstrate the superiority of the proposed SeqWalker.

Baichen Liu, Qi Lyu, Xudong Wang et al.

Continual video instance segmentation demands both the plasticity to absorb new object categories and the stability to retain previously learned ones, all while preserving temporal consistency across frames. In this work, we introduce Contrastive Residual Injection and Semantic Prompting (CRISP), an earlier attempt tailored to address the instance-wise, category-wise, and task-wise confusion in continual video instance segmentation. For instance-wise learning, we model instance tracking and construct instance correlation loss, which emphasizes the correlation with the prior query space while strengthening the specificity of the current task query. For category-wise learning, we build an adaptive residual semantic prompt (ARSP) learning framework, which constructs a learnable semantic residual prompt pool generated by category text and uses an adjustive query-prompt matching mechanism to build a mapping relationship between the query of the current task and the semantic residual prompt. Meanwhile, a semantic consistency loss based on the contrastive learning is introduced to maintain semantic coherence between object queries and residual prompts during incremental training. For task-wise learning, to ensure the correlation at the inter-task level within the query space, we introduce a concise yet powerful initialization strategy for incremental prompts. Extensive experiments on YouTube-VIS-2019 and YouTube-VIS-2021 datasets demonstrate that CRISP significantly outperforms existing continual segmentation methods in the long-term continual video instance segmentation task, avoiding catastrophic forgetting and effectively improving segmentation and classification performance. The code is available at https://github.com/01upup10/CRISP.

Shiben Liu, Mingyue Xu, Huijie Fan et al.

Lifelong person re-identification (LReID) encounters a key challenge: balancing the preservation of old knowledge with adaptation to new information. Existing LReID methods typically employ knowledge distillation to enforce representation alignment. However, these approaches ignore two crucial aspects: specific distribution awareness and cross-domain unified knowledge learning, both of which are essential for addressing this challenge. To overcome these limitations, we propose a novel distribution-aware knowledge unification and association (DKUA) framework where domain-style modeling is performed for each instance to propagate domain-specific representations, enhancing anti-forgetting and generalization capacity. Specifically, we design a distribution-aware model to transfer instance-level representations of the current domain into the domain-specific representations with the different domain styles, preserving learned knowledge without storing old samples. Next, we propose adaptive knowledge consolidation (AKC) to dynamically generate the unified representation as a cross-domain representation center. To further mitigate forgetting, we develop a unified knowledge association (UKA) mechanism, which explores the unified representation as a bridge to explicitly model inter-domain associations, reducing inter-domain gaps. Finally, distribution-based knowledge transfer (DKT) is proposed to prevent the current domain distribution from deviating from the cross-domain distribution center, improving adaptation capacity. Experimental results show our DKUA outperforms the existing methods by 7.6%/5.3% average mAP/R@1 improvement on anti-forgetting and generalization capacity, respectively. Our code is available at https://github.com/LiuShiBen/DKUA.

Siyuan Wang, Jiawei Liu, Wei Wang et al.

Co-Speech Gesture Video Generation aims to generate vivid speech videos from audio-driven still images, which is challenging due to the diversity of different parts of the body in terms of amplitude of motion, audio relevance, and detailed features. Relying solely on audio as the control signal often fails to capture large gesture movements in video, leading to more pronounced artifacts and distortions. Existing approaches typically address this issue by introducing additional a priori information, but this can limit the practical application of the task. Specifically, we propose a Motion Mask-Guided Two-Stage Network (MMGT) that uses audio, as well as motion masks and motion features generated from the audio signal to jointly drive the generation of synchronized speech gesture videos. In the first stage, the Spatial Mask-Guided Audio Pose Generation (SMGA) Network generates high-quality pose videos and motion masks from audio, effectively capturing large movements in key regions such as the face and gestures. In the second stage, we integrate the Motion Masked Hierarchical Audio Attention (MM-HAA) into the Stabilized Diffusion Video Generation model, overcoming limitations in fine-grained motion generation and region-specific detail control found in traditional methods. This guarantees high-quality, detailed upper-body video generation with accurate texture and motion details. Evaluations show improved video quality, lip-sync, and gesture. The model and code are available at https://github.com/SIA-IDE/MMGT.

Xing Xie, Jiawei Liu, Ziyue Lin et al.

We present Autoregressive Representation Alignment (ARRA), a new training framework that unlocks global-coherent text-to-image generation in autoregressive LLMs without architectural modifications. Different from prior works that require complex architectural redesigns, ARRA aligns LLM's hidden states with visual representations from external visual foundational models via a global visual alignment loss and a hybrid token, <HYBNEXT>. This token enforces dual constraints: local next-token prediction and global semantic distillation, enabling LLMs to implicitly learn spatial and contextual coherence while retaining their original autoregressive paradigm. Extensive experiments validate ARRA's plug-and-play versatility. When training T2I LLMs from scratch, ARRA reduces FID by 16.6% (ImageNet), 12.0% (LAION-COCO) for autoregressive LLMs like LlamaGen, without modifying original architecture and inference mechanism. For training from text-generation-only LLMs, ARRA reduces FID by 25.5% (MIMIC-CXR), 8.8% (DeepEyeNet) for advanced LLMs like Chameleon. For domain adaptation, ARRA aligns general-purpose LLMs with specialized models (e.g., BioMedCLIP), achieving an 18.6% FID reduction over direct fine-tuning on medical imaging (MIMIC-CXR). These results demonstrate that training objective redesign, rather than architectural modifications, can resolve cross-modal global coherence challenges. ARRA offers a complementary paradigm for advancing autoregressive models. The code is available at https://github.com/HKU-HealthAI/ARRA.

Xudong Wang, Jiahua Dong, Baichen Liu et al.

Embodied navigation agents powered by large language models have shown strong performance on individual tasks but struggle to continually acquire new navigation skills, which suffer from catastrophic forgetting. We formalize this challenge as lifelong embodied navigation learning (LENL), where an agent is required to adapt to a sequence of navigation tasks spanning multiple scenes and diverse user instruction styles, while retaining previously learned knowledge. To tackle this problem, we propose Uni-Walker, a lifelong embodied navigation framework that decouples navigation knowledge into task-shared and task-specific components with Decoder Extension LoRA (DE-LoRA). To learn the shared knowledge, we design a knowledge inheritance strategy and an experts co-activation strategy to facilitate shared knowledge transfer and refinement across multiple navigation tasks. To learn the specific knowledge, we propose an expert subspace orthogonality constraint together and a navigation-specific chain-of-thought reasoning mechanism to capture specific knowledge and enhance instruction-style understanding. Extensive experiments demonstrate the superiority of Uni-Walker for building universal navigation agents with lifelong learning.

Shiben Liu, Huijie Fan, Qiang Wang et al.

Lifelong Person Re-Identification (LReID) aims to continuously learn from successive data streams, matching individuals across multiple cameras. The key challenge for LReID is how to effectively preserve old knowledge while incrementally learning new information, which is caused by task-level domain gaps and limited old task datasets. Existing methods based on CNN backbone are insufficient to explore the representation of each instance from different perspectives, limiting model performance on limited old task datasets and new task datasets. Unlike these methods, we propose a Diverse Representations Embedding (DRE) framework that first explores a pure transformer for LReID. The proposed DRE preserves old knowledge while adapting to new information based on instance-level and task-level layout. Concretely, an Adaptive Constraint Module (ACM) is proposed to implement integration and push away operations between multiple overlapping representations generated by transformer-based backbone, obtaining rich and discriminative representations for each instance to improve adaptive ability of LReID. Based on the processed diverse representations, we propose Knowledge Update (KU) and Knowledge Preservation (KP) strategies at the task-level layout by introducing the adjustment model and the learner model. KU strategy enhances the adaptive learning ability of learner models for new information under the adjustment model prior, and KP strategy preserves old knowledge operated by representation-level alignment and logit-level supervision in limited old task datasets while guaranteeing the adaptive learning information capacity of the LReID model. Compared to state-of-the-art methods, our method achieves significantly improved performance in holistic, large-scale, and occluded datasets.

Zhiyu Liu, Zhi Han, Yandong Tang et al.

This paper considers the problem of recovering a tensor with an underlying low-tubal-rank structure from a small number of corrupted linear measurements. Traditional approaches tackling such a problem require the computation of tensor Singular Value Decomposition (t-SVD), that is a computationally intensive process, rendering them impractical for dealing with large-scale tensors. Aim to address this challenge, we propose an efficient and effective low-tubal-rank tensor recovery method based on a factorization procedure akin to the Burer-Monteiro (BM) method. Precisely, our fundamental approach involves decomposing a large tensor into two smaller factor tensors, followed by solving the problem through factorized gradient descent (FGD). This strategy eliminates the need for t-SVD computation, thereby reducing computational costs and storage requirements. We provide rigorous theoretical analysis to ensure the convergence of FGD under both noise-free and noisy situations. Additionally, it is worth noting that our method does not require the precise estimation of the tensor tubal-rank. Even in cases where the tubal-rank is slightly overestimated, our approach continues to demonstrate robust performance. A series of experiments have been carried out to demonstrate that, as compared to other popular ones, our approach exhibits superior performance in multiple scenarios, in terms of the faster computational speed and the smaller convergence error.

Zhiyu Liu, Zhi Han, Yandong Tang et al.

We consider the noisy matrix sensing problem in the over-parameterization setting, where the estimated rank $r$ is larger than the true rank $r_\star$ of the target matrix $X_\star$. Specifically, our main objective is to recover a matrix $ X_\star \in \mathbb{R}^{n_1 \times n_2} $ with rank $ r_\star $ from noisy measurements using an over-parameterized factorization $ LR^\top $, where $ L \in \mathbb{R}^{n_1 \times r}, \, R \in \mathbb{R}^{n_2 \times r} $ and $ \min\{n_1, n_2\} \ge r > r_\star $, with $ r_\star $ being unknown. Recently, preconditioning methods have been proposed to accelerate the convergence of matrix sensing problem compared to vanilla gradient descent, incorporating preconditioning terms $ (L^\top L + λI)^{-1} $ and $ (R^\top R + λI)^{-1} $ into the original gradient. However, these methods require careful tuning of the damping parameter $λ$ and are sensitive to step size. To address these limitations, we propose the alternating preconditioned gradient descent (APGD) algorithm, which alternately updates the two factor matrices, eliminating the need for the damping parameter $λ$ and enabling faster convergence with larger step sizes. We theoretically prove that APGD convergences to a near-optimal error at a linear rate. We further show that APGD can be extended to deal with other low-rank matrix estimation tasks, also with a theoretical guarantee of linear convergence. To validate the effectiveness and scalability of the proposed APGD, we conduct simulated and real-world experiments on a wide range of low-rank estimation problems, including noisy matrix sensing, weighted PCA, 1-bit matrix completion, and matrix completion. The extensive results demonstrate that APGD consistently achieves the fastest convergence and the lowest computation time compared to the existing alternatives.

Xing Xie, Jiawei Liu, Huijie Fan et al.

Directly reconstructing 3D CT volume from few-view 2D X-rays using an end-to-end deep learning network is a challenging task, as X-ray images are merely projection views of the 3D CT volume. In this work, we facilitate complex 2D X-ray image to 3D CT mapping by incorporating new view synthesis, and reduce the learning difficulty through view-guided feature alignment. Specifically, we propose a dual-view guided diffusion model (DVG-Diffusion), which couples a real input X-ray view and a synthesized new X-ray view to jointly guide CT reconstruction. First, a novel view parameter-guided encoder captures features from X-rays that are spatially aligned with CT. Next, we concatenate the extracted dual-view features as conditions for the latent diffusion model to learn and refine the CT latent representation. Finally, the CT latent representation is decoded into a CT volume in pixel space. By incorporating view parameter guided encoding and dual-view guided CT reconstruction, our DVG-Diffusion can achieve an effective balance between high fidelity and perceptual quality for CT reconstruction. Experimental results demonstrate our method outperforms state-of-the-art methods. Based on experiments, the comprehensive analysis and discussions for views and reconstruction are also presented.

Linyao Chen, Bo Huang, Qinlao Zhao et al.

Agentic Web, as a new paradigm that redefines the internet through autonomous, goal-driven interactions, plays an important role in group intelligence. As the foundational semantic primitives of the Agentic Web, digital assets encapsulate interactive web elements into agents, which expand the capacities and coverage of agents in agentic web. The lack of automated methodologies for agent generation limits the wider usage of digital assets and the advancement of the Agentic Web. In this paper, we first formalize these challenges by strictly defining the A2A-Agentization process, decomposing it into critical stages and identifying key technical hurdles on top of the A2A protocol. Based on this framework, we develop an Agentization Agent to agentize digital assets for the Agentic Web. To rigorously evaluate this capability, we propose A2A-Agentization Bench, the first benchmark explicitly designed to evaluate agentization quality in terms of fidelity and interoperability. Our experiments demonstrate that our approach effectively activates the functional capabilities of digital assets and enables interoperable A2A multi-agent collaboration. We believe this work will further facilitate scalable and standardized integration of digital assets into the Agentic Web ecosystem.

Xudong Wang, Zebin Han, Zhiyu Liu et al.

Traditional language-conditioned manipulation agent sequential adaptation to new manipulation skills leads to catastrophic forgetting of old skills, limiting dynamic scene practical deployment. In this paper, we propose SkillsCrafter, a novel robotic manipulation framework designed to continually learn multiple skills while reducing catastrophic forgetting of old skills. Specifically, we propose a Manipulation Skills Adaptation to retain the old skills knowledge while inheriting the shared knowledge between new and old skills to facilitate learning of new skills. Meanwhile, we perform the singular value decomposition on the diverse skill instructions to obtain common skill semantic subspace projection matrices, thereby recording the essential semantic space of skills. To achieve forget-less and generalization manipulation, we propose a Skills Specialization Aggregation to compute inter-skills similarity in skill semantic subspaces, achieving aggregation of the previously learned skill knowledge for any new or unknown skill. Extensive experiments demonstrate the effectiveness and superiority of our proposed SkillsCrafter.

Yanmei Wang, Xiyao Liu, Fupeng Chu et al.

Domain generalization aims at training on source domains to uncover a domain-invariant feature space, allowing the model to perform robust generalization ability on unknown target domains. However, due to domain gaps, it is hard to find reliable common image feature space, and the reason for that is the lack of suitable basic units for images. Different from image in vision space, language has comprehensive expression elements that can effectively convey semantics. Inspired by the semantic completeness of language and intuitiveness of image, we propose VLCA, which combine language space and vision space, and connect the multiple image domains by using semantic space as the bridge domain. Specifically, in language space, by taking advantage of the completeness of language basic units, we tend to capture the semantic representation of the relations between categories through word vector distance. Then, in vision space, by taking advantage of the intuitiveness of image features, the common pattern of sample features with the same class is explored through low-rank approximation. In the end, the language representation is aligned with the vision representation through the multimodal space of text and image. Experiments demonstrate the effectiveness of the proposed method.

Jingyang Qiao, Zhizhong Zhang, Xin Tan et al.

Parameter-efficient tunings (PETs) have demonstrated impressive performance and promising perspectives in training large models, while they are still confronted with a common problem: the trade-off between learning new content and protecting old knowledge, leading to zero-shot generalization collapse, and cross-modal hallucination. In this paper, we reformulate Adapter, LoRA, Prefix-tuning, and Prompt-tuning from the perspective of gradient projection, and firstly propose a unified framework called Parameter Efficient Gradient Projection (PEGP). We introduce orthogonal gradient projection into different PET paradigms and theoretically demonstrate that the orthogonal condition for the gradient can effectively resist forgetting even for large-scale models. It therefore modifies the gradient towards the direction that has less impact on the old feature space, with less extra memory space and training time. We extensively evaluate our method with different backbones, including ViT and CLIP, on diverse datasets, and experiments comprehensively demonstrate its efficiency in reducing forgetting in class, online class, domain, task, and multi-modality continual settings. The project page is available at https://dmcv-ecnu-pegp.github.io/.

Yang Zhang, Yao Wang, Zhi Han et al.

In recent years, there have been an increasing number of applications of tensor completion based on the tensor train (TT) format because of its efficiency and effectiveness in dealing with higher-order tensor data. However, existing tensor completion methods using TT decomposition have two obvious drawbacks. One is that they only consider mode weights according to the degree of mode balance, even though some elements are recovered better in an unbalanced mode. The other is that serious blocking artifacts appear when the missing element rate is relatively large. To remedy such two issues, in this work, we propose a novel tensor completion approach via the element-wise weighted technique. Accordingly, a novel formulation for tensor completion and an effective optimization algorithm, called as tensor completion by parallel weighted matrix factorization via tensor train (TWMac-TT), is proposed. In addition, we specifically consider the recovery quality of edge elements from adjacent blocks. Different from traditional reshaping and ket augmentation, we utilize a new tensor augmentation technique called overlapping ket augmentation, which can further avoid blocking artifacts. We then conduct extensive performance evaluations on synthetic data and several real image data sets. Our experimental results demonstrate that the proposed algorithm TWMac-TT outperforms several other competing tensor completion methods.

Zhi Han, Siquan Yu, Shao-Bo Lin et al.

Deep learning is recognized to be capable of discovering deep features for representation learning and pattern recognition without requiring elegant feature engineering techniques by taking advantage of human ingenuity and prior knowledge. Thus it has triggered enormous research activities in machine learning and pattern recognition. One of the most important challenge of deep learning is to figure out relations between a feature and the depth of deep neural networks (deep nets for short) to reflect the necessity of depth. Our purpose is to quantify this feature-depth correspondence in feature extraction and generalization. We present the adaptivity of features to depths and vice-verse via showing a depth-parameter trade-off in extracting both single feature and composite features. Based on these results, we prove that implementing the classical empirical risk minimization on deep nets can achieve the optimal generalization performance for numerous learning tasks. Our theoretical results are verified by a series of numerical experiments including toy simulations and a real application of earthquake seismic intensity prediction.

Xi'ai Chen, Zhi Han, Yao Wang et al.

Because of the limitations of matrix factorization, such as losing spatial structure information, the concept of low-rank tensor factorization (LRTF) has been applied for the recovery of a low dimensional subspace from high dimensional visual data. The low-rank tensor recovery is generally achieved by minimizing the loss function between the observed data and the factorization representation. The loss function is designed in various forms under different noise distribution assumptions, like $L_1$ norm for Laplacian distribution and $L_2$ norm for Gaussian distribution. However, they often fail to tackle the real data which are corrupted by the noise with unknown distribution. In this paper, we propose a generalized weighted low-rank tensor factorization method (GWLRTF) integrated with the idea of noise modelling. This procedure treats the target data as high-order tensor directly and models the noise by a Mixture of Gaussians, which is called MoG GWLRTF. The parameters in the model are estimated under the EM framework and through a new developed algorithm of weighted low-rank tensor factorization. We provide two versions of the algorithm with different tensor factorization operations, i.e., CP factorization and Tucker factorization. Extensive experiments indicate the respective advantages of this two versions in different applications and also demonstrate the effectiveness of MoG GWLRTF compared with other competing methods.

Shiying He, Haiwei Zhou, Yao Wang et al.

In this paper, we propose a novel approach to hyperspectral image super-resolution by modeling the global spatial-and-spectral correlation and local smoothness properties over hyperspectral images. Specifically, we utilize the tensor nuclear norm and tensor folded-concave penalty functions to describe the global spatial-and-spectral correlation hidden in hyperspectral images, and 3D total variation (TV) to characterize the local spatial-and-spectral smoothness across all hyperspectral bands. Then, we develop an efficient algorithm for solving the resulting optimization problem by combing the local linear approximation (LLA) strategy and alternative direction method of multipliers (ADMM). Experimental results on one hyperspectral image dataset illustrate the merits of the proposed approach.

Zhi Han, Zongben Xu, Song-Chun Zhu

This paper presents a middle-level video representation named Video Primal Sketch (VPS), which integrates two regimes of models: i) sparse coding model using static or moving primitives to explicitly represent moving corners, lines, feature points, etc., ii) FRAME /MRF model reproducing feature statistics extracted from input video to implicitly represent textured motion, such as water and fire. The feature statistics include histograms of spatio-temporal filters and velocity distributions. This paper makes three contributions to the literature: i) Learning a dictionary of video primitives using parametric generative models; ii) Proposing the Spatio-Temporal FRAME (ST-FRAME) and Motion-Appearance FRAME (MA-FRAME) models for modeling and synthesizing textured motion; and iii) Developing a parsimonious hybrid model for generic video representation. Given an input video, VPS selects the proper models automatically for different motion patterns and is compatible with high-level action representations. In the experiments, we synthesize a number of textured motion; reconstruct real videos using the VPS; report a series of human perception experiments to verify the quality of reconstructed videos; demonstrate how the VPS changes over the scale transition in videos; and present the close connection between VPS and high-level action models.

Liangqiong Qu, Jiandong Tian, Zhi Han et al.

In this paper, we propose a novel, effective and fast method to obtain a color illumination invariant and shadow-free image from a single outdoor image. Different from state-of-the-art methods for shadow-free image that either need shadow detection or statistical learning, we set up a linear equation set for each pixel value vector based on physically-based shadow invariants, deduce a pixel-wise orthogonal decomposition for its solutions, and then get an illumination invariant vector for each pixel value vector on an image. The illumination invariant vector is the unique particular solution of the linear equation set, which is orthogonal to its free solutions. With this illumination invariant vector and Lab color space, we propose an algorithm to generate a shadow-free image which well preserves the texture and color information of the original image. A series of experiments on a diverse set of outdoor images and the comparisons with the state-of-the-art methods validate our method.