arXiv subjects

Yali Wang

Publications and source records attributed to Yali Wang.

At least 37 records · Page 2Linked to original sources

arXiv2026

UniFlow: A Unified Pixel Flow Tokenizer for Visual Understanding and Generation

Tokenizer is a crucial component for both visual understanding and generation. To advance toward the ultimate goal of universal modeling, recent research has focused on developing a unified tokenizer. However, existing tokenizers face a significant performance trade-off between understanding and generation, stemming from the inherent conflict between high-level semantic abstraction and low-level pixel reconstruction. To tackle this challenge, we propose a generic and unified tokenizer, namely UniFlow, by flexibly adapting any visual encoder with a concise reconstruction decoder. Specifically, we introduce layer-wise adaptive self-distillation applied to the well-pretrained visual encoders, which enables UniFlow to simultaneously inherit the strong semantic features for visual understanding and flexibly adapt to model fine-grained details for visual generation. Moreover, we propose a lightweight patch-wise pixel flow decoder, which efficiently achieves high-fidelity pixel reconstruction by modeling a conditional flow from the noisy state back to the patch-wise pixel domain. By leveraging the semantic features as visual conditions for the decoder, we effectively alleviate the training conflicts between understanding and generation. Furthermore, the patch-wise learning strategy simplifies the data distribution, thereby improving training efficiency. Extensive experiments across 13 challenging benchmarks spanning 7 widely studied visual understanding and generation tasks demonstrate that UniFlow achieves a win-win outcome. For instance, our 7B UniFlow-XL not only surpasses the 14B TokenFlow-XL by 6.05% on average understanding benchmarks, but also achieves a competitive results in both visual reconstruction and generation, surpassing UniTok by 0.15 in rFID and 0.09 in gFID (without guidance), respectively.

cs.CV↗

arXiv2026

MotionWeaver: Holistic 4D-Anchored Framework for Multi-Humanoid Image Animation

Character image animation, which synthesizes videos of reference characters driven by pose sequences, has advanced rapidly but remains largely limited to single-human settings. Existing methods struggle to generalize to multi-humanoid scenarios, which involve diverse humanoid forms, complex interactions, and frequent occlusions. We address this gap with two key innovations. First, we introduce unified motion representations that extract identity-agnostic motions and explicitly bind them to corresponding characters, enabling generalization across diverse humanoid forms and seamless extension to multi-humanoid scenarios. Second, we propose a holistic 4D-anchored paradigm that constructs a shared 4D space to fuse motion representations with video latents, and further reinforces this process with hierarchical 4D-level supervision to better handle interactions and occlusions. We instantiate these ideas in MotionWeaver, an end-to-end framework for multi-humanoid image animation. To support this setting, we curate a 46-hour dataset of multi-human videos with rich interactions, and construct a 300-video benchmark featuring paired humanoid characters. Quantitative and qualitative experiments demonstrate that MotionWeaver not only achieves state-of-the-art results on our benchmark but also generalizes effectively across diverse humanoid forms, complex interactions, and challenging multi-humanoid scenarios.

cs.CV↗

arXiv2026

WeTok: Powerful Discrete Tokenization for High-Fidelity Visual Reconstruction

Visual tokenizer is a critical component for vision generation. However, the existing tokenizers often face unsatisfactory trade-off between compression ratios and reconstruction fidelity. To fill this gap, we introduce a powerful and concise WeTok tokenizer, which surpasses the previous leading tokenizers via two core innovations. (1) Group-wise lookup-free Quantization (GQ). We partition the latent features into groups, and perform lookup-free quantization for each group. As a result, GQ can efficiently overcome memory and computation limitations of prior tokenizers, while achieving a reconstruction breakthrough with more scalable codebooks. (2) Generative Decoder (GD). Different from prior tokenizers, we introduce a generative decoder with a prior of extra noise variable. In this case, GD can probabilistically model the distribution of visual data conditioned on discrete tokens, allowing WeTok to reconstruct visual details, especially at high compression ratio. On the ImageNet 50k validation set, at a high-fidelity setting, WeTok achieves a record-low zero-shot rFID of 0.12, outperforming leading continuous tokenizers like FLUX-VAE (0.18) and SD-VAE 3.5 (0.19) with 400% compression ratio. Furthermore, in a high-compression regime, WeTok achieves a zero-shot rFID of 3.49 at a 768$\times$ compression ratio, substantially surpassing Cosmos, which scores 4.57 at only 50% our compression ratio. Code and models are available: https://github.com/zhuangshaobin/WeTok.

cs.CV↗

arXiv2025

Super Encoding Network: Recursive Association of Multi-Modal Encoders for Video Understanding

Video understanding has been considered as one critical step towards world modeling, which is an important long-term problem in AI research. Recently, multimodal foundation models have shown such potential via large-scale pretraining. These models effectively align encoders of different modalities via contrastive learning. To further enhance performance on complex target movements and diversified video scenes, we propose to augment this alignment with deeper multimodal interactions, which are critical for understanding complex target movements with diversified video scenes. To fill this gap, we propose a unified Super Encoding Network (SEN) for video understanding, which builds up such distinct interactions through the recursive association of multimodal encoders in the foundation models. Specifically, we creatively treat those well-trained encoders as ``super neurons" in our SEN. Via designing a Recursive Association (RA) block, we progressively fuse multi-modalities with the input video, based on knowledge integrating, distributing, and prompting of super neurons in a recursive manner. In this way, our SEN can effectively encode deeper multimodal interactions for prompting various video understanding tasks in the downstream. Extensive experiments show that our SEN can remarkably boost the four most representative video tasks, including tracking, recognition, chatting, and editing, e.g., for pixel-level tracking, the average jaccard index improves 2.7%, and temporal coherence(TC) drops by 8.8% compared to the popular CaDeX++ approach. For one-shot video editing, textual alignment improves 6.4%, and frame consistency increases by 4.1% compared to the Tune-A-Video approach.

cs.CV↗

arXiv2025

LVAgent: Long Video Understanding by Multi-Round Dynamical Collaboration of MLLM Agents

Existing MLLMs encounter significant challenges in modeling the temporal context within long videos. Currently, mainstream Agent-based methods use external tools to assist a single MLLM in answering long video questions. Despite such tool-based support, a solitary MLLM still offers only a partial understanding of long videos, resulting in limited performance. In order to better address long video tasks, we introduce LVAgent, the first framework enabling multi-round dynamic collaboration of MLLM agents in long video understanding. Our method consists of four key steps: 1) Selection: We pre-select appropriate agents from the model library to form optimal agent teams based on different tasks. 2) Perception: We design an effective retrieval scheme for long videos to improve the coverage of critical temporal segments while maintaining computational efficiency. 3) Action: Agents answer long video questions and exchange reasons. 4) Reflection: We evaluate each agent's performance in each round of discussion and optimize the agent team for dynamic collaboration. The agents iteratively refine their answers by multi-round dynamical collaboration of MLLM agents. LVAgent is the first agent system method that outperforms all closed-source models (like GPT-4o) and open-source models (like InternVL-2.5 and Qwen2-VL) in the long video understanding tasks. Our LVAgent achieves an accuracy of 80\% on four mainstream long video understanding tasks. Notably, LVAgent improves accuracy by 13.3\% on LongVideoBench. Code is available at https://github.com/64327069/LVAgent.

cs.CV↗

arXiv2025

Low-Resolution Action Recognition for Tiny Actions Challenge

Tiny Actions Challenge focuses on understanding human activities in real-world surveillance. Basically, there are two main difficulties for activity recognition in this scenario. First, human activities are often recorded at a distance, and appear in a small resolution without much discriminative clue. Second, these activities are naturally distributed in a long-tailed way. It is hard to alleviate data bias for such heavy category imbalance. To tackle these problems, we propose a comprehensive recognition solution in this paper. First, we train video backbones with data balance, in order to alleviate overfitting in the challenge benchmark. Second, we design a dual-resolution distillation framework, which can effectively guide low-resolution action recognition by super-resolution knowledge. Finally, we apply model en-semble with post-processing, which can further boost per-formance on the long-tailed categories. Our solution ranks Top-1 on the leaderboard.

cs.CV↗

arXiv2025

Percept, Chat, and then Adapt: Multimodal Knowledge Transfer of Foundation Models for Open-World Video Recognition

Open-world video recognition is challenging since traditional networks are not generalized well on complex environment variations. Alternatively, foundation models with rich knowledge have recently shown their generalization power. However, how to apply such knowledge has not been fully explored for open-world video recognition. To this end, we propose a generic knowledge transfer pipeline, which progressively exploits and integrates external multimodal knowledge from foundation models to boost open-world video recognition. We name it PCA, based on three stages of Percept, Chat, and Adapt. First, we perform Percept process to reduce the video domain gap and obtain external visual knowledge. Second, we generate rich linguistic semantics as external textual knowledge in Chat stage. Finally, we blend external multimodal knowledge in Adapt stage, by inserting multimodal knowledge adaptation modules into networks. We conduct extensive experiments on three challenging open-world video benchmarks, i.e., TinyVIRAT, ARID, and QV-Pipe. Our approach achieves state-of-the-art performance on all three datasets.

cs.CV↗

arXiv2025

G-UBS: Towards Robust Understanding of Implicit Feedback via Group-Aware User Behavior Simulation

User feedback is critical for refining recommendation systems, yet explicit feedback (e.g., likes or dislikes) remains scarce in practice. As a more feasible alternative, inferring user preferences from massive implicit feedback has shown great potential (e.g., a user quickly skipping a recommended video usually indicates disinterest). Unfortunately, implicit feedback is often noisy: a user might skip a video due to accidental clicks or other reasons, rather than disliking it. Such noise can easily misjudge user interests, thereby undermining recommendation performance. To address this issue, we propose a novel Group-aware User Behavior Simulation (G-UBS) paradigm, which leverages contextual guidance from relevant user groups, enabling robust and in-depth interpretation of implicit feedback for individual users. Specifically, G-UBS operates via two key agents. First, the User Group Manager (UGM) effectively clusters users to generate group profiles utilizing a ``summarize-cluster-reflect" workflow based on LLMs. Second, the User Feedback Modeler (UFM) employs an innovative group-aware reinforcement learning approach, where each user is guided by the associated group profiles during the reinforcement learning process, allowing UFM to robustly and deeply examine the reasons behind implicit feedback. To assess our G-UBS paradigm, we have constructed a Video Recommendation benchmark with Implicit Feedback (IF-VR). To the best of our knowledge, this is the first multi-modal benchmark for implicit feedback evaluation in video recommendation, encompassing 15k users, 25k videos, and 933k interaction records with implicit feedback. Extensive experiments on IF-VR demonstrate that G-UBS significantly outperforms mainstream LLMs and MLLMs, with a 4.0% higher proportion of videos achieving a play rate > 30% and 14.9% higher reasoning accuracy on IF-VR.

cs.IR↗

arXiv2025

When Top-ranked Recommendations Fail: Modeling Multi-Granular Negative Feedback for Explainable and Robust Video Recommendation

Existing video recommendation systems, relying mainly on ID-based embedding mapping and collaborative filtering, often fail to capture in-depth video content semantics. Moreover, most struggle to address biased user behaviors (e.g., accidental clicks, fast skips), leading to inaccurate interest modeling and frequent negative feedback in top recommendations with unclear causes. To tackle this issue, we collect real-world user video-watching sequences, annotate the reasons for users' dislikes, and construct a benchmark dataset for personalized explanations. We then introduce the Agentic Explainable Negative Feedback (ENF) framework, which integrates three core components: (1) the Profile Agent, extracting behavioral cues from users' historical data to derive psychological and personality profiles; (2) the Video Agent, performing comprehensive multimodal video analysis; and (3) the Reason Agent, synthesizing information from the other two agents to predict user engagement and generate explanations. Additionally, we propose the S-GRPO algorithm, enabling the model to progressively address complex tasks during reinforcement fine-tuning. Experimental results on the collected dataset show that our method significantly outperforms state-of-the-art baselines in negative feedback prediction and reason explanation. Notably, it achieves an 8.6% improvement over GPT-4o in reason classification. Deployment on the business platform further validates its benefits: increasing average user watch time by 6.2%, reducing the fast-skip rate by 9.4%, and significantly enhancing user satisfaction.

cs.MM↗

arXiv2025

VideoChat-R1: Enhancing Spatio-Temporal Perception via Reinforcement Fine-Tuning

Reinforcement Learning (RL) benefits Large Language Models (LLMs) for complex reasoning. Inspired by this, we explore integrating spatio-temporal specific rewards into Multimodal Large Language Models (MLLMs) to address the unique challenges of video understanding, such as long-range temporal associations. This paper investigates how rule-based rewards, particularly temporal ones, can improve video reasoning and their generalizability. Our study proposes Reinforcement Fine-Tuning (RFT) as a data-efficient method to enhance video reasoning on specific tasks without sacrificing original capabilities. Through joint RFT on multiple spatio-temporal perception tasks, we developed VideoChat-R1, a powerful Video MLLM. VideoChat-R1 achieves state-of-the-art spatio-temporal perception, demonstrating significant improvements in tasks like temporal grounding (+31.8) and object tracking (+31.2), while also improving general QA benchmarks. The enhanced perception and preserved chat abilities contribute to a more reliable video dialogue system, leading to our ``Temporal Clue-driven Reasoning" inference schema. This work provides a foundation for developing robust, real-world video comprehension agents.

cs.CV↗

arXiv2025

VideoTG-R1: Boosting Video Temporal Grounding via Curriculum Reinforcement Learning on Reflected Boundary Annotations

Video temporal grounding (VTG) aims to locate precise segments in videos based on language queries, which is a fundamental challenge in video understanding. While recent Multimodal Large Language Models (MLLMs) have shown promise in tackling VTG through reinforcement learning (RL), they overlook the challenges arising from both the quality and difficulty of training samples. (1) Partially annotated samples. Many samples contain relevant segments beyond the annotated interval, introducing ambiguous supervision. (2) Hard-to-ground samples. Samples with poor zero-shot performance produce consistently low and indistinguishable rewards during RL training, exhibiting no clear preference among multiple outputs and thus hindering learning efficiency. To address these challenges, we propose VideoTG-R1, a novel curriculum RL framework with reflected boundary annotations, enabling data-efficient training. Specifically, we propose a Boundary Reflection Agent that utilizes MLLMs to predict query-relevant timestamps outside the annotated intervals, allowing us to identify and filter out partially annotated samples, thereby reducing ambiguity. Furthermore, we introduce a Difficulty Estimation Agent to assess the training difficulty of each sample and design a curriculum RL strategy that dynamically masks the videos of hard-to-ground samples according to the training steps, easing the training difficulty and providing clearer preference. Experiments on the VTG and grounded VideoQA tasks demonstrate the effectiveness of our method. Remarkably, with only 10% of the training samples and 21% of the computational budget, VideoTG-R1 outperforms full-data counterparts under both group relative policy optimization (GRPO) and supervised fine-tuning (SFT). The code is available at https://github.com/ldong1111/VideoTG-R1.

cs.CV↗

arXiv2025

A Renaissance of Explicit Motion Information Mining from Transformers for Action Recognition

Recently, action recognition has been dominated by transformer-based methods, thanks to their spatiotemporal contextual aggregation capacities. However, despite the significant progress achieved on scene-related datasets, they do not perform well on motion-sensitive datasets due to the lack of elaborate motion modeling designs. Meanwhile, we observe that the widely-used cost volume in traditional action recognition is highly similar to the affinity matrix defined in self-attention, but equipped with powerful motion modeling capacities. In light of this, we propose to integrate those effective motion modeling properties into the existing transformer in a unified and neat way, with the proposal of the Explicit Motion Information Mining module (EMIM). In EMIM, we propose to construct the desirable affinity matrix in a cost volume style, where the set of key candidate tokens is sampled from the query-based neighboring area in the next frame in a sliding-window manner. Then, the constructed affinity matrix is used to aggregate contextual information for appearance modeling and is converted into motion features for motion modeling as well. We validate the motion modeling capacities of our method on four widely-used datasets, and our method performs better than existing state-of-the-art approaches, especially on motion-sensitive datasets, i.e., Something-Something V1 & V2. Our project is available at https://github.com/PeiqinZhuang/EMIM .

cs.CV↗

arXiv2025

Beyond Textual CoT: Interleaved Text-Image Chains with Deep Confidence Reasoning for Image Editing

Image editing with natural language has gained significant popularity, yet existing methods struggle with intricate object intersections and fine-grained spatial relationships due to the lack of an explicit reasoning process. While Chain-of-Thought (CoT) has been explored to enhance reasoning, purely textual CoT or CoT augmented with coordinate information is fundamentally limited in its ability to represent intricate visual layouts and lacks the necessary visual cues to guide the generation of fine-grained, pixel-level details. To address these challenges, we propose Multimodal Reasoning Edit (MURE), a novel framework that shifts the visual editing process from purely text-based reasoning to a series of interleaved textual and visual rationales. Our framework performs image editing using a natively multimodal, interleaved text-image CoT. This approach generates a step-by-step chain of reasoning where a textual description is followed by a corresponding visual cue, such as a positional mask that defined intended edited regions or a representation of new content. Furthermore, to mitigate the hallucination phenomenon of large language models, we introduce Multimodal Deep Confidence (MMDC) reasoning paradigm. This paradigm explores a tree of visual reasoning paths at each step. By pruning low-quality branches using a deep confidence score from a reward model, it ensures the model consistently follows a high-quality trajectory towards the final edited result. The proposed method decomposes complex editing tasks into interdependent sub-tasks, achieving greater precision at each stage and yielding high-fidelity edited results. We define the formulation for interleaved text-image chains and release the first CoT-Edit-14K dataset, comprising 14K high-quality editing examples. Extensive experiments show that our method yields significant improvements across three image editing benchmarks.

cs.CV↗

arXiv2025

VideoChat-R1.5: Visual Test-Time Scaling to Reinforce Multimodal Reasoning by Iterative Perception

Inducing reasoning in multimodal large language models (MLLMs) is critical for achieving human-level perception and understanding. Existing methods mainly leverage LLM reasoning to analyze parsed visuals, often limited by static perception stages. This paper introduces Visual Test-Time Scaling (VTTS), a novel approach to enhance MLLMs' reasoning via iterative perception during inference. VTTS mimics humans' hierarchical attention by progressively refining focus on high-confidence spatio-temporal regions, guided by updated textual predictions. Specifically, VTTS employs an Iterative Perception (ITP) mechanism, incorporating reinforcement learning with spatio-temporal supervision to optimize reasoning. To support this paradigm, we also present VTTS-80K, a dataset tailored for iterative perception. These designs allows a MLLM to enhance its performance by increasing its perceptual compute. Extensive experiments validate VTTS's effectiveness and generalization across diverse tasks and benchmarks. Our newly introduced Videochat-R1.5 model has achieved remarkable improvements, with an average increase of over 5\%, compared to robust baselines such as Qwen2.5VL-3B and -7B, across more than 15 benchmarks that encompass video conversation, video reasoning, and spatio-temporal perception.

cs.CV↗

arXiv2025

LvBench: A Benchmark for Long-form Video Understanding with Versatile Multi-modal Question Answering

Despite remarkable recent progress, existing long-form VideoQA datasets fall short of meeting the criteria for genuine long-form video understanding. This is primarily due to the use of short videos for question curation, and the reliance on limited-length sub-clips as clues to answer those questions. Meanwhile, previous datasets have limited focus on question type and modality. To remedy this, we introduce LvBench, a Long-form video understanding benchmark for versatile multi-modal question-answering. Our LvBench stands out from existing long-form VideoQA datasets through three key characteristics: 1) Extended temporal durations: We consider videos ranging from 70 seconds to 4 hours, covering single-scene, multi-scene, and full-scene contexts. This design accounts for both video and clue lengths, capturing diverse contextual dynamics. 2) Diverse question types and modalities: LvBench introduces six distinct question types that evaluate various perceptual and cognitive capabilities, utilizing both video frames and subtitles. 3) High-quality annotations: We employ rigorous manual labeling by human annotators. Our dataset comprises 20,061 question-answer pairs sourced from 100 carefully selected movies across diverse genres, annotated collaboratively by multiple individuals. Analysis involving various baselines reveals a consistent trend: the performance of all existing methods significantly deteriorates when video and clue length increases. We expect LvBench to serve as a valuable resource for future works on long-form video understanding.

cs.CV↗

arXiv2025

VRBench: A Benchmark for Multi-Step Reasoning in Long Narrative Videos

We present VRBench, the first long narrative video benchmark crafted for evaluating large models' multi-step reasoning capabilities, addressing limitations in existing evaluations that overlook temporal reasoning and procedural validity. It comprises 960 long videos (with an average duration of 1.6 hours), along with 8,243 human-labeled multi-step question-answering pairs and 25,106 reasoning steps with timestamps. These videos are curated via a multi-stage filtering process including expert inter-rater reviewing to prioritize plot coherence. We develop a human-AI collaborative framework that generates coherent reasoning chains, each requiring multiple temporally grounded steps, spanning seven types (e.g., event attribution, implicit inference). VRBench designs a multi-phase evaluation pipeline that assesses models at both the outcome and process levels. Apart from the MCQs for the final results, we propose a progress-level LLM-guided scoring metric to evaluate the quality of the reasoning chain from multiple dimensions comprehensively. Through extensive evaluations of 12 LLMs and 19 VLMs on VRBench, we undertake a thorough analysis and provide valuable insights that advance the field of multi-step reasoning.

cs.CV↗

arXiv2025

VideoChat-Flash: Hierarchical Compression for Long-Context Video Modeling

Long-context video modeling is critical for multimodal large language models (MLLMs), enabling them to process movies, online video streams, and so on. Despite its advances, handling long videos remains challenging due to the difficulty in efficiently understanding the extremely long video context. This paper aims to address this issue from aspects of model architecture, training data, training strategy and evaluation benchmark. First, we propose a novel Hierarchical video token Compression (HiCo) method, which leverages visual redundancy in long videos to compress long video context from Clip-level to Video-level, reducing the computation significantly while preserving essential details, achieving an extreme compression ratio of approximately 1/50 with almost no performance loss. Second, we introduce a multi-stage short-to-long learning scheme, a large-scale dataset of real-world long videos named LongVid, and a challenging ``Multi-Hop Needle-In-A-Video-Haystack'' benchmark. Finally, we build a powerful video MLLM named VideoChat-Flash, which shows a leading performance on both mainstream long and short video benchmarks at the 2B and 7B model scale. It first gets 99.1% accuracy over 10,000 frames in NIAH among open-source models.

cs.CV↗

arXiv2025

InternVideo2.5: Empowering Video MLLMs with Long and Rich Context Modeling

This paper aims to improve the performance of video multimodal large language models (MLLM) via long and rich context (LRC) modeling. As a result, we develop a new version of InternVideo2.5 with a focus on enhancing the original MLLMs' ability to perceive fine-grained details and capture long-form temporal structure in videos. Specifically, our approach incorporates dense vision task annotations into MLLMs using direct preference optimization and develops compact spatiotemporal representations through adaptive hierarchical token compression. Experimental results demonstrate this unique design of LRC greatly improves the results of video MLLM in mainstream video understanding benchmarks (short & long), enabling the MLLM to memorize significantly longer video inputs (at least 6x longer than the original), and master specialized vision capabilities like object tracking and segmentation. Our work highlights the importance of multimodal context richness (length and fineness) in empowering MLLM's innate abilites (focus and memory), providing new insights for future research on video MLLM. Code and models are available at https://github.com/OpenGVLab/InternVideo/tree/main/InternVideo2.5

cs.CV↗