Zhucun Xue

Jiangning Zhang, Haoyang He, Xuhai Chen et al.

Large Multimodal Model (LMM) GPT-4V(ision) endows GPT-4 with visual grounding capabilities, making it possible to handle certain tasks through the Visual Question Answering (VQA) paradigm. This paper explores the potential of VQA-oriented GPT-4V in the recently popular visual Anomaly Detection (AD) and is the first to conduct qualitative and quantitative evaluations on the popular MVTec AD and VisA datasets. Considering that this task requires both image-/pixel-level evaluations, the proposed GPT-4V-AD framework contains three components: \textbf{\textit{1)}} Granular Region Division, \textbf{\textit{2)}} Prompt Designing, \textbf{\textit{3)}} Text2Segmentation for easy quantitative evaluation, and have made some different attempts for comparative analysis. The results show that GPT-4V can achieve certain results in the zero-shot AD task through a VQA paradigm, such as achieving image-level 77.1/88.0 and pixel-level 68.0/76.6 AU-ROCs on MVTec AD and VisA datasets, respectively. However, its performance still has a certain gap compared to the state-of-the-art zero-shot method, \eg, WinCLIP and CLIP-AD, and further researches are needed. This study provides a baseline reference for the research of VQA-oriented LMM in the zero-shot AD task, and we also post several possible future works. Code is available at \url{https://github.com/zhangzjn/GPT-4V-AD}.

Yinan Chen, Chuming Lin, Zhennan Chen et al.

While instruction-based video editing has seen significant progress, joint audio-visual editing remains constrained by the absence of dedicated datasets and benchmarks. To bridge this gap, we present JAVEdit-100k, the first large-scale, high-quality dataset tailored for instruction-guided joint audio-visual editing. Focusing on human-centric videos, JAVEdit-100k comprises approximately 100K editing triplets spanning five distinct categories, including subject editing and speech editing. This dataset is rigorously constructed via four meticulously designed generation pipelines, seamlessly paired with an agent-in-the-loop quality control mechanism. Furthermore, to address the lack of standardized evaluation within the field, we introduce JAVEditBench, a comprehensive benchmark featuring curated source videos and human-aligned instructions across all editing categories. Finally, we propose JAVEdit, a pioneering baseline model for instruction-guided joint audio-visual editing. Experiments show that \model\ outperforms all baselines on five of six evaluation metrics.

Jiangning Zhang, Xiangtai Li, Jian Li et al.

This paper focuses on developing modern, efficient, lightweight models for dense predictions while trading off parameters, FLOPs, and performance. Inverted Residual Block (IRB) serves as the infrastructure for lightweight CNNs, but no counterpart has been recognized by attention-based studies. This work rethinks lightweight infrastructure from efficient IRB and effective components of Transformer from a unified perspective, extending CNN-based IRB to attention-based models and abstracting a one-residual Meta Mobile Block (MMB) for lightweight model design. Following simple but effective design criterion, we deduce a modern Inverted Residual Mobile Block (iRMB) and build a ResNet-like Efficient MOdel (EMO) with only iRMB for down-stream tasks. Extensive experiments on ImageNet-1K, COCO2017, and ADE20K benchmarks demonstrate the superiority of our EMO over state-of-the-art methods, e.g., EMO-1M/2M/5M achieve 71.5, 75.1, and 78.4 Top-1 that surpass equal-order CNN-/Attention-based models, while trading-off the parameter, efficiency, and accuracy well: running 2.8-4.0x faster than EdgeNeXt on iPhone14.

Tong Zhang, Jiangning Zhang, Zhucun Xue et al.

Balancing convergence speed, generalization capability, and computational efficiency remains a core challenge in deep learning optimization. First-order gradient descent methods, epitomized by stochastic gradient descent (SGD) and Adam, serve as the cornerstone of modern training pipelines. However, large-scale model training, stringent differential privacy requirements, and distributed learning paradigms expose critical limitations in these conventional approaches regarding privacy protection and memory efficiency. To mitigate these bottlenecks, researchers explore second-order optimization techniques to surpass first-order performance ceilings, while zeroth-order methods reemerge to alleviate memory constraints inherent to large-scale training. Despite this proliferation of methodologies, the field lacks a cohesive framework that unifies underlying principles and delineates application scenarios for these disparate approaches. In this work, we retrospectively analyze the evolutionary trajectory of deep learning optimization algorithms and present a comprehensive empirical evaluation of mainstream optimizers across diverse model architectures and training scenarios. We distill key emerging trends and fundamental design trade-offs, pinpointing promising directions for future research. By synthesizing theoretical insights with extensive empirical evidence, we provide actionable guidance for designing next-generation highly efficient, robust, and trustworthy optimization methods. The code is available at https://github.com/APRIL-AIGC/Awesome-Optimizer.

Haojun Chen, Haoyang He, Chengming Xu et al.

Text-to-Image (T2I) models have recently seen notable progress around 1K and 2K resolution. With the extreme desire for better visual experience and the rapid development of imaging technology, the demand for Ultra-High-Resolution (UHR) image generation has grown significantly. However, UHR image generation poses great challenges due to the scarcity and complexity of high-resolution content. In this paper, we first introduce PixVerve-95K, a high-quality, open-source UHR T2I dataset curated with a carefully designed data pipeline, which contains 95K images across diverse scenarios (each image has a minimum pixel-count of 100M) and seven-dimensional annotations. Based on our large-scale image-text dataset, we take a pioneering step to extend various T2I foundation models to native 100MP generation with three training schemes. Finally, leveraging both conventional metrics and multimodal large language model-based assessments, our proposed PixVerve-Bench benchmark establishes a comprehensive evaluation protocol for UHR images encompassing visual quality and semantic alignment. Extensive experimental results on our benchmark and the constructive exploration of training strategies collaboratively provide valuable insights for future breakthroughs.

Yicheng Xu, Jiangning Zhang, Zhucun Xue et al.

In-context Learning enables training-free adaptation via demonstrations but remains highly sensitive to example selection and formatting. In unified multimodal models spanning understanding and generation, this sensitivity is exacerbated by cross-modal interference and varying cognitive demands. Consequently, In-context Learning efficacy is often non-monotonic and highly task-dependent. To diagnose these behaviors, we introduce a six-level capability-oriented taxonomy that categorizes the functional role of demonstrations from basic perception to high-order discernment. Guided by this cognitive framework, we construct UniICL-760K, a large-scale corpus featuring curated 8-shot In-context Learning episodes across 15 subtasks, alongside UniICL-Bench for rigorous, controlled evaluation. As an architectural intervention to stabilize few-shot adaptation, we propose the Context-Adaptive Prototype Modulator, a lightweight, plug-and-play module. Evaluations on UniICL-Bench show that our approach yields highly competitive unified results, outperforming larger-parameter multimodal large language model baselines on most understanding In-context Learning tasks. Data and code will be available soon at https://github.com/xuyicheng-zju/UniICL.

Teng Hu, Jiangning Zhang, Hongrui Huang et al.

The rapid advancement of Artificial Intelligence Generated Content (AIGC) has revolutionized video generation, enabling systems ranging from proprietary pioneers like OpenAI's Sora, Google's Veo3, and Bytedance's Seedance to powerful open-source contenders like Wan and HunyuanVideo to synthesize temporally coherent and semantically rich videos. These advancements pave the way for building "world models" that simulate real-world dynamics, with applications spanning entertainment, education, and virtual reality. However, existing reviews on video generation often focus on narrow technical fields, e.g., Generative Adversarial Networks (GAN) and diffusion models, or specific tasks (e. g., video editing), lacking a comprehensive perspective on the field's evolution, especially regarding Auto-Regressive (AR) models and integration of multimodal information. To address these gaps, this survey firstly provides a systematic review of the development of video generation technology, tracing its evolution from early GANs to dominant diffusion models, and further to emerging AR-based and multimodal techniques. We conduct an in-depth analysis of the foundational principles, key advancements, and comparative strengths/limitations. Then, we explore emerging trends in multimodal video generation, emphasizing the integration of diverse data types to enhance contextual awareness. Finally, by bridging historical developments and contemporary innovations, this survey offers insights to guide future research in video generation and its applications, including virtual/augmented reality, personalized education, autonomous driving simulations, digital entertainment, and advanced world models, in this rapidly evolving field. For more details, please refer to the project at https://github.com/sjtuplayer/Awesome-Video-Foundations.

Linxuan Fan, Juntao Jiang, Weixuan Liu et al.

Skin lesion segmentation is a crucial step in dermatology for guiding clinical decision-making. However, existing methods for accurate, robust, and resource-efficient lesion analysis have limitations, including low performance and high computational complexity. To address these limitations, we propose UltraLBM-UNet, a lightweight U-Net variant that integrates a bidirectional Mamba-based global modeling mechanism with multi-branch local feature perception. The proposed architecture integrates efficient local feature injection with bidirectional state-space modeling, enabling richer contextual interaction across spatial dimensions while maintaining computational compactness suitable for point-of-care deployment. Extensive experiments on the ISIC 2017, ISIC 2018, and PH2 datasets demonstrate that our model consistently achieves state-of-the-art segmentation accuracy, outperforming existing lightweight and Mamba counterparts with only 0.034M parameters and 0.060 GFLOPs. In addition, we introduce a hybrid knowledge distillation strategy to train an ultra-compact student model, where the distilled variant UltraLBM-UNet-T, with only 0.011M parameters and 0.019 GFLOPs, achieves competitive segmentation performance. These results highlight the suitability of UltraLBM-UNet for point-of-care deployment, where accurate and robust lesion analyses are essential. The source code is publicly available at https://github.com/LinLinLin-X/UltraLBM-UNet.

Zhucun Xue, Jiangning Zhang, Juntao Jiang et al.

The rapid expansion of research across machine learning, vision, and language has produced a volume of publications that is increasingly difficult to synthesize. Traditional bibliometric tools rely mainly on metadata and offer limited visibility into the semantic content of papers, making it hard to track how research themes evolve over time or how different areas influence one another. To obtain a clearer picture of recent developments, we compile a unified corpus of more than 100,000 papers from 22 major conferences between 2020 and 2025 and construct a multidimensional profiling pipeline to organize and analyze their textual content. By combining topic clustering, LLM-assisted parsing, and structured retrieval, we derive a comprehensive representation of research activity that supports the study of topic lifecycles, methodological transitions, dataset and model usage patterns, and institutional research directions. Our analysis highlights several notable shifts, including the growth of safety, multimodal reasoning, and agent-oriented studies, as well as the gradual stabilization of areas such as neural machine translation and graph-based methods. These findings provide an evidence-based view of how AI research is evolving and offer a resource for understanding broader trends and identifying emerging directions. Code and dataset: https://github.com/xzc-zju/Profiling_Scientific_Literature

Haoyang He, Jie Wang, Jiangning Zhang et al.

The quality and diversity of instruction-based image editing datasets are continuously increasing, yet large-scale, high-quality datasets for instruction-based video editing remain scarce. To address this gap, we introduce OpenVE-3M, an open-source, large-scale, and high-quality dataset for instruction-based video editing. It comprises two primary categories: spatially-aligned edits (Global Style, Background Change, Local Change, Local Remove, Local Add, and Subtitles Edit) and non-spatially-aligned edits (Camera Multi-Shot Edit and Creative Edit). All edit types are generated via a meticulously designed data pipeline with rigorous quality filtering. OpenVE-3M surpasses existing open-source datasets in terms of scale, diversity of edit types, instruction length, and overall quality. Furthermore, to address the lack of a unified benchmark in the field, we construct OpenVE-Bench, containing 431 video-edit pairs that cover a diverse range of editing tasks with three key metrics highly aligned with human judgment. We present OpenVE-Edit, a 5B model trained on our dataset that demonstrates remarkable efficiency and effectiveness by setting a new state-of-the-art on OpenVE-Bench, outperforming all prior open-source models including a 14B baseline. Project page is at https://github.com/lewandofskee/OpenVE.

Yabiao Wang, Shuo Wang, Jiangning Zhang et al.

Human-human motion generation is essential for understanding humans as social beings. Current methods fall into two main categories: single-person-based methods and separate modeling-based methods. To delve into this field, we abstract the overall generation process into a general framework MetaMotion, which consists of two phases: temporal modeling and interaction mixing. For temporal modeling, the single-person-based methods concatenate two people into a single one directly, while the separate modeling-based methods skip the modeling of interaction sequences. The inadequate modeling described above resulted in sub-optimal performance and redundant model parameters. In this paper, we introduce TIMotion (Temporal and Interactive Modeling), an efficient and effective framework for human-human motion generation. Specifically, we first propose Causal Interactive Injection to model two separate sequences as a causal sequence leveraging the temporal and causal properties. Then we present Role-Evolving Scanning to adjust to the change in the active and passive roles throughout the interaction. Finally, to generate smoother and more rational motion, we design Localized Pattern Amplification to capture short-term motion patterns. Extensive experiments on InterHuman and InterX demonstrate that our method achieves superior performance. Project page: https://aigc-explorer.github.io/TIMotion-page/

Yue Han, Junwei Zhu, Yuxiang Feng et al.

Current diffusion-based face animation methods generally adopt a ReferenceNet (a copy of U-Net) and a large amount of curated self-acquired data to learn appearance features, as robust appearance features are vital for ensuring temporal stability. However, when trained on public datasets, the results often exhibit a noticeable performance gap in image quality and temporal consistency. To address this issue, we meticulously examine the essential appearance features in the facial animation tasks, which include motion-agnostic (e.g., clothing, background) and motion-related (e.g., facial details) texture components, along with high-level discriminative identity features. Drawing from this analysis, we introduce a Motion-Identity Modulated Appearance Learning Module (MIA) that modulates CLIP features at both motion and identity levels. Additionally, to tackle the semantic/ color discontinuities between clips, we design an Inter-clip Affinity Learning Module (ICA) to model temporal relationships across clips. Our method achieves precise facial motion control (i.e., expressions and gaze), faithful identity preservation, and generates animation videos that maintain both intra/inter-clip temporal consistency. Moreover, it easily adapts to various modalities of driving sources. Extensive experiments demonstrate the superiority of our method.

Yuxuan Cai, Jiangning Zhang, Haoyang He et al.

The success of Large Language Models (LLMs) has inspired the development of Multimodal Large Language Models (MLLMs) for unified understanding of vision and language. However, the increasing model size and computational complexity of large-scale MLLMs (l-MLLMs) limit their use in resource-constrained scenarios. Although small-scale MLLMs (s-MLLMs) are designed to reduce computational costs, they typically suffer from performance degradation. To mitigate this limitation, we propose a novel LLaVA-KD framework to transfer knowledge from l-MLLMs to s-MLLMs. Specifically, we introduce Multimodal Distillation (MDist) to transfer teacher model's robust representations across both visual and linguistic modalities, and Relation Distillation (RDist) to transfer teacher model's ability to capture visual token relationships. Additionally, we propose a three-stage training scheme to fully exploit the potential of the proposed distillation strategy: 1) Distilled Pre-Training to strengthen the alignment between visual-linguistic representations in s-MLLMs, 2) Supervised Fine-Tuning to equip the s-MLLMs with multimodal understanding capacity, and 3) Distilled Fine-Tuning to refine s-MLLM's knowledge. Our approach significantly improves s-MLLMs performance without altering the model architecture. Extensive experiments and ablation studies validate the effectiveness of each proposed component. Code will be available at https://github.com/Fantasyele/LLaVA-KD.

Jiangning Zhang, Junwei Zhu, Teng Hu et al.

Native 4K (2160$\times$3840) video generation remains a critical challenge due to the quadratic computational explosion of full-attention as spatiotemporal resolution increases, making it difficult for models to strike a balance between efficiency and quality. This paper proposes a novel Transformer retrofit strategy termed $\textbf{T3}$ ($\textbf{T}$ransform $\textbf{T}$rained $\textbf{T}$ransformer) that, without altering the core architecture of full-attention pretrained models, significantly reduces compute requirements by optimizing their forward logic. Specifically, $\textbf{T3-Video}$ introduces a multi-scale weight-sharing window attention mechanism and, via hierarchical blocking together with an axis-preserving full-attention design, can effect an "attention pattern" transformation of a pretrained model using only modest compute and data. Results on 4K-VBench show that $\textbf{T3-Video}$ substantially outperforms existing approaches: while delivering performance improvements (+4.29$\uparrow$ VQA and +0.08$\uparrow$ VTC), it accelerates native 4K video generation by more than 10$\times$. Project page at https://zhangzjn.github.io/projects/T3-Video

Jiangning Zhang, Chengjie Wang, Xiangtai Li et al.

Anomaly detection (AD) is often focused on detecting anomaly areas for industrial quality inspection and medical lesion examination. However, due to the specific scenario targets, the data scale for AD is relatively small, and evaluation metrics are still deficient compared to classic vision tasks, such as object detection and semantic segmentation. To fill these gaps, this work first constructs a large-scale and general-purpose COCO-AD dataset by extending COCO to the AD field. This enables fair evaluation and sustainable development for different methods on this challenging benchmark. Moreover, current metrics such as AU-ROC have nearly reached saturation on simple datasets, which prevents a comprehensive evaluation of different methods. Inspired by the metrics in the segmentation field, we further propose several more practical threshold-dependent AD-specific metrics, ie, m$F_1$$^{.2}_{.8}$, mAcc$^{.2}_{.8}$, mIoU$^{.2}_{.8}$, and mIoU-max. Motivated by GAN inversion's high-quality reconstruction capability, we propose a simple but more powerful InvAD framework to achieve high-quality feature reconstruction. Our method improves the effectiveness of reconstruction-based methods on popular MVTec AD, VisA, and our newly proposed COCO-AD datasets under a multi-class unsupervised setting, where only a single detection model is trained to detect anomalies from different classes. Extensive ablation experiments have demonstrated the effectiveness of each component of our InvAD. Full codes and models are available at https://github.com/zhangzjn/ader.

Jinzhuo Liu, Jiangning Zhang, Wencan Jiang et al.

Autoregressive video generation has improved rapidly in visual fidelity and interactivity, but it still suffers from long-term inconsistency and memory degradation. Most existing solutions either compress historical frames using predefined strategies or retrieve keyframes based on coarse implicit attention signals, both of which fail to handle evolving prompts with shifting entity references, leading to identity drift, character duplication, and attribute loss. To address this, we propose IAMFlow, a training-free identity-aware memory framework that explicitly models and tracks persistent entity identities, enabling consistent generation across prompt transitions. Specifically, an LLM extracts entities with visual attributes from each prompt and assigns unique global IDs for identity-aware memory, while a VLM asynchronously verifies and refines attributes from rendered frames, enabling explicit entity tracking in place of implicit similarity-based matching. To keep the proposed framework computationally practical, we design a systematic inference acceleration pipeline, including asynchronous visual verification, adaptive prompt transition, and model quantization, which achieves faster generation than existing baselines. Furthermore, we introduce NarraStream-Bench, a benchmark for narrative streaming video generation that features 324 multi-prompt scripts spanning six dimensions and a three-dimensional evaluation protocol that integrates both traditional metrics and multimodal large language model-based assessments. Extensive experiments show that IAMFlow, despite being training-free, achieves the best overall performance on NarraStream-Bench, outperforming the strongest baseline by 2.56 points, while achieving a 1.39$\times$ speedup over the most efficient baseline in the 60-second multi-prompt setting.

Wencan Jiang, Jiangning Zhang, Jianbiao Mei et al.

Long-horizon multimodal agents in open-world games must stay goal-directed across many low-level interactions under tight token and latency budgets. Existing approaches often trade off costly per-step reasoning against reactive execution that can drift, repeat failures, and recover poorly. Our key idea is to reuse strategic reasoning across locally stable segments and reinvoke it at event boundaries. We present SPIKE, an adaptive dual controller framework for cost-efficient long-horizon game control. Its Strategic Controller performs low-frequency global planning, failure analysis, and recovery, while its Reactive Controller handles fast local execution under a strict token budget. An Event Trigger monitors visual change, task progress, repeated actions, and failure signals to decide when control should stay reactive or escalate to strategic reasoning. Hierarchical Memory separates short-term experience reuse in the State-Action Memory Bank (SA-MB) from structured evidence in the State Action Knowledge Graph (SA-KG), allowing each controller to retrieve the context it needs. This design reuses strategic proposals over multiple reactive steps, supports local override when plans become stale, and reserves expensive reasoning for moments where extra deliberation is useful. On the Lite-100 split of StarDojo, SPIKE improves Lite-100 success rate (SR) by 5.0 percentage points (38.5% relative) over the strongest Lite-100 baseline and Budgeted SR by 9.3 points (75.6% relative) over the strongest budgeted baseline. It also reduces token consumption by 54.9% and latency by 40.8%. Ablations show that event triggering, reactive override, and heterogeneous memory each contribute to success and recovery, supporting selective reasoning rather than reasoning at every step.

Juntao Jiang, Jiangning Zhang, Yali Bi et al.

Chain-of-Thought (CoT) reasoning has proven effective in enhancing large language models by encouraging step-by-step intermediate reasoning, and recent advances have extended this paradigm to Multimodal Large Language Models (MLLMs). In the medical domain, where diagnostic decisions depend on nuanced visual cues and sequential reasoning, CoT aligns naturally with clinical thinking processes. However, Current benchmarks for medical image understanding generally focus on the final answer while ignoring the reasoning path. An opaque process lacks reliable bases for judgment, making it difficult to assist doctors in diagnosis. To address this gap, we introduce a new M3CoTBench benchmark specifically designed to evaluate the correctness, efficiency, impact, and consistency of CoT reasoning in medical image understanding. M3CoTBench features 1) a diverse, multi-level difficulty dataset covering 24 examination types, 2) 13 varying-difficulty tasks, 3) a suite of CoT-specific evaluation metrics (correctness, efficiency, impact, and consistency) tailored to clinical reasoning, and 4) a performance analysis of multiple MLLMs. M3CoTBench systematically evaluates CoT reasoning across diverse medical imaging tasks, revealing current limitations of MLLMs in generating reliable and clinically interpretable reasoning, and aims to foster the development of transparent, trustworthy, and diagnostically accurate AI systems for healthcare. Project page at https://juntaojianggavin.github.io/projects/M3CoTBench/.

Zhucun Xue, Jiangning Zhang, Teng Hu et al.

The quality of the video dataset (image quality, resolution, and fine-grained caption) greatly influences the performance of the video generation model. The growing demand for video applications sets higher requirements for high-quality video generation models. For example, the generation of movie-level Ultra-High Definition (UHD) videos and the creation of 4K short video content. However, the existing public datasets cannot support related research and applications. In this paper, we first propose a high-quality open-sourced UHD-4K (22.4\% of which are 8K) text-to-video dataset named UltraVideo, which contains a wide range of topics (more than 100 kinds), and each video has 9 structured captions with one summarized caption (average of 824 words). Specifically, we carefully design a highly automated curation process with four stages to obtain the final high-quality dataset: \textit{i)} collection of diverse and high-quality video clips. \textit{ii)} statistical data filtering. \textit{iii)} model-based data purification. \textit{iv)} generation of comprehensive, structured captions. In addition, we expand Wan to UltraWan-1K/-4K, which can natively generate high-quality 1K/4K videos with more consistent text controllability, demonstrating the effectiveness of our data curation.We believe that this work can make a significant contribution to future research on UHD video generation. UltraVideo dataset and UltraWan models are available at https://xzc-zju.github.io/projects/UltraVideo.

Teng Hu, Jiangning Zhang, Ran Yi et al.

Employing LLMs for visual generation has recently become a research focus. However, the existing methods primarily transfer the LLM architecture to visual generation but rarely investigate the fundamental differences between language and vision. This oversight may lead to suboptimal utilization of visual generation capabilities within the LLM framework. In this paper, we explore the characteristics of visual embedding space under the LLM framework and discover that the correlation between visual embeddings can help achieve more stable and robust generation results. We present IAR, an Improved AutoRegressive Visual Generation Method that enhances the training efficiency and generation quality of LLM-based visual generation models. Firstly, we propose a Codebook Rearrangement strategy that uses balanced k-means clustering algorithm to rearrange the visual codebook into clusters, ensuring high similarity among visual features within each cluster. Leveraging the rearranged codebook, we propose a Cluster-oriented Cross-entropy Loss that guides the model to correctly predict the cluster where the token is located. This approach ensures that even if the model predicts the wrong token index, there is a high probability the predicted token is located in the correct cluster, which significantly enhances the generation quality and robustness. Extensive experiments demonstrate that our method consistently enhances the model training efficiency and performance from 100M to 1.4B, reducing the training time by half while achieving the same FID. Additionally, our approach can be applied to various LLM-based visual generation models and adheres to the scaling law, providing a promising direction for future research in LLM-based visual generation. The code is available at: https://github.com/sjtuplayer/IAR.

Zhucun Xue, Jiangning Zhang, Xurong Xie et al.

Multimodal Large Language Models (MLLMs) struggle with long videos due to fixed context windows and weak long-term dependency modeling. Existing Retrieval-Augmented Generation (RAG) methods for videos use static retrieval strategies, leading to inefficiencies for simple queries and information loss for complex tasks. To address this, we propose AdaVideoRAG, a novel framework that dynamically adapts retrieval granularity based on query complexity using a lightweight intent classifier. Our framework employs an Omni-Knowledge Indexing module to build hierarchical databases from text (captions, ASR, OCR), visual features, and semantic graphs, enabling optimal resource allocation across tasks. We also introduce the HiVU benchmark for comprehensive evaluation. Experiments demonstrate improved efficiency and accuracy for long-video understanding, with seamless integration into existing MLLMs. AdaVideoRAG establishes a new paradigm for adaptive retrieval in video analysis. Codes will be open-sourced at https://github.com/xzc-zju/AdaVideoRAG.

Jiangning Zhang, Teng Hu, Haoyang He et al.

This work focuses on developing parameter-efficient and lightweight models for dense predictions while trading off parameters, FLOPs, and performance. Our goal is to set up the new frontier of the 5M magnitude lightweight model on various downstream tasks. Inverted Residual Block (IRB) serves as the infrastructure for lightweight CNNs, but no counterparts have been recognized by attention-based design. Our work rethinks the lightweight infrastructure of efficient IRB and practical components in Transformer from a unified perspective, extending CNN-based IRB to attention-based models and abstracting a one-residual Meta Mobile Block (MMBlock) for lightweight model design. Following neat but effective design criterion, we deduce a modern Improved Inverted Residual Mobile Block (i2RMB) and improve a hierarchical Efficient MOdel (EMOv2) with no elaborate complex structures. Considering the imperceptible latency for mobile users when downloading models under 4G/5G bandwidth and ensuring model performance, we investigate the performance upper limit of lightweight models with a magnitude of 5M. Extensive experiments on various vision recognition, dense prediction, and image generation tasks demonstrate the superiority of our EMOv2 over state-of-the-art methods, e.g., EMOv2-1M/2M/5M achieve 72.3, 75.8, and 79.4 Top-1 that surpass equal-order CNN-/Attention-based models significantly. At the same time, EMOv2-5M equipped RetinaNet achieves 41.5 mAP for object detection tasks that surpasses the previous EMO-5M by +2.6. When employing the more robust training recipe, our EMOv2-5M eventually achieves 82.9 Top-1 accuracy, which elevates the performance of 5M magnitude models to a new level. Code is available at https://github.com/zhangzjn/EMOv2.

Xinlei Yu, Chengming Xu, Guibin Zhang et al.

Multi-Agent System (MAS) powered by Visual Language Models (VLMs) enables challenging tasks but suffers from a novel failure term, multi-agent visual hallucination snowballing, where hallucinations are seeded in a single agent and amplified by following ones due to the over-reliance on textual flow to relay visual information. Through turn-, layer-, and token-wise attention analyses, we provide detailed insights into the essence of hallucination snowballing regarding the reduction of visual attention allocation. It leads us to identify a subset of vision tokens with a unimodal attention peak in middle layers that best preserve visual evidence but gradually diminish in deeper agent turns, resulting in the visual hallucination snowballing in MAS. Thus, we propose ViF, a lightweight, plug-and-play mitigation paradigm that relays inter-agent messages with Visual Flow powered by the selected visual relay tokens and applies attention reallocation to amplify this pattern. The experiment results demonstrate that our method markedly reduces hallucination snowballing, consistently improving the performance across eight benchmarks based on four common MAS structures and ten base models. The source code is publicly available at: https://github.com/YU-deep/ViF.git.

Jiangning Zhang, Haoyang He, Zhenye Gan et al.

Visual anomaly detection aims to identify anomalous regions in images through unsupervised learning paradigms, with increasing application demand and value in fields such as industrial inspection and medical lesion detection. Despite significant progress in recent years, there is a lack of comprehensive benchmarks to adequately evaluate the performance of various mainstream methods across different datasets under the practical multi-class setting. The absence of standardized experimental setups can lead to potential biases in training epochs, resolution, and metric results, resulting in erroneous conclusions. This paper addresses this issue by proposing a comprehensive visual anomaly detection benchmark, ADer, which is a modular framework that is highly extensible for new methods. The benchmark includes multiple datasets from industrial and medical domains, implementing fifteen state-of-the-art methods and nine comprehensive metrics. Additionally, we have proposed the GPU-assisted ADEval package to address the slow evaluation problem of metrics like time-consuming mAU-PRO on large-scale data, significantly reducing evaluation time by more than 1000-fold. Through extensive experimental results, we objectively reveal the strengths and weaknesses of different methods and provide insights into the challenges and future directions of multi-class visual anomaly detection. We hope that ADer will become a valuable resource for researchers and practitioners in the field, promoting the development of more robust and generalizable anomaly detection systems. Full codes are open-sourced at https://github.com/zhangzjn/ader.

Yinan Chen, Jiangning Zhang, Yali Bi et al.

Image inversion is a fundamental task in generative models, aiming to map images back to their latent representations to enable downstream applications such as editing, restoration, and style transfer. This paper provides a comprehensive review of the latest advancements in image inversion techniques, focusing on two main paradigms: Generative Adversarial Network (GAN) inversion and diffusion model inversion. We categorize these techniques based on their optimization methods. For GAN inversion, we systematically classify existing methods into encoder-based approaches, latent optimization approaches, and hybrid approaches, analyzing their theoretical foundations, technical innovations, and practical trade-offs. For diffusion model inversion, we explore training-free strategies, fine-tuning methods, and the design of additional trainable modules, highlighting their unique advantages and limitations. Additionally, we discuss several popular downstream applications and emerging applications beyond image tasks, identifying current challenges and future research directions. By synthesizing the latest developments, this paper aims to provide researchers and practitioners with a valuable reference resource, promoting further advancements in the field of image inversion. We keep track of the latest works at https://github.com/RyanChenYN/ImageInversion

Juntao Jiang, Jiangning Zhang, Weixuan Liu et al.

In recent years, significant advancements have been made in deep learning for medical image segmentation, particularly with convolutional neural networks (CNNs) and transformer models. However, CNNs face limitations in capturing long-range dependencies, while transformers suffer from high computational complexity. To address this, we propose RWKV-UNet, a novel model that integrates the RWKV (Receptance Weighted Key Value) structure into the U-Net architecture. This integration enhances the model's ability to capture long-range dependencies and to improve contextual understanding, which is crucial for accurate medical image segmentation. We build a strong encoder with developed Global-Local Spatial Perception (GLSP) blocks combining CNNs and RWKVs. We also propose a Cross-Channel Mix (CCM) module to improve skip connections with multi-scale feature fusion, achieving global channel information integration. Experiments on 11 benchmark datasets show that the RWKV-UNet achieves state-of-the-art performance on various types of medical image segmentation tasks. Additionally, smaller variants, RWKV-UNet-S and RWKV-UNet-T, balance accuracy and computational efficiency, making them suitable for broader clinical applications.

Yinan Chen, Jiangning Zhang, Teng Hu et al.

Instruction-guided video editing has emerged as a rapidly advancing research direction, offering new opportunities for intuitive content transformation while also posing significant challenges for systematic evaluation. Existing video editing benchmarks fail to support the evaluation of instruction-guided video editing adequately and further suffer from limited source diversity, narrow task coverage and incomplete evaluation metrics. To address the above limitations, we introduce IVEBench, a modern benchmark suite specifically designed for instruction-guided video editing assessment. IVEBench comprises a diverse database of 600 high-quality source videos, spanning seven semantic dimensions, and covering video lengths ranging from 32 to 1,024 frames. It further includes 8 categories of editing tasks with 35 subcategories, whose prompts are generated and refined through large language models and expert review. Crucially, IVEBench establishes a three-dimensional evaluation protocol encompassing video quality, instruction compliance and video fidelity, integrating both traditional metrics and multimodal large language model-based assessments. Extensive experiments demonstrate the effectiveness of IVEBench in benchmarking state-of-the-art instruction-guided video editing methods, showing its ability to provide comprehensive and human-aligned evaluation outcomes.

Yuhang Dong, Haizhou Ge, Yupei Zeng et al.

Visuomotor imitation learning policies enable robots to efficiently acquire manipulation skills from visual demonstrations. However, as scene complexity and visual distractions increase, policies that perform well in simple settings often experience substantial performance degradation. To address this challenge, we propose ImitDiff, a diffusion-based imitation learning policy guided by fine-grained semantics within a dual-resolution workflow. Leveraging pretrained priors of vision-language foundation models, our method transforms high-level instructions into pixel-level visual semantic masks. These masks guide a dual-resolution perception pipeline that captures both global context (e.g., overall layout) from low-resolution observation and fine-grained local features (e.g., geometric details) from high-resolution observation, enabling the policy to focus on task-relevant regions. Additionally, we introduce a consistency-driven diffusion transformer action head that bridges visual semantic conditions and real-time action generation. Extensive experiments demonstrate that ImitDiff outperforms state-of-the-art vision-language manipulation frameworks, as well as visuomotor imitation learning policies, particularly under increased scene complexity and visual distractions. Notably, ImitDiff exhibits strong generalization in zero-shot settings involving novel objects and visual distractions. Furthermore, our consistency-driven action head achieves an order-of-magnitude improvement in inference speed while maintaining competitive success rates.

Dongqi Liu, Hang Ding, Qiming Feng et al.

Retrieval-Augmented Generation (RAG) has emerged as an important means of enhancing the performance of large language models (LLMs) in knowledge-intensive tasks. However, most existing RAG strategies treat retrieved passages in a flat and unstructured way, which prevents the model from capturing structural cues and constrains its ability to synthesize knowledge from dispersed evidence across documents. To overcome these limitations, we propose Disco-RAG, a discourse-aware framework that explicitly injects discourse signals into the generation process. Our method constructs intra-chunk discourse trees to capture local hierarchies and builds inter-chunk rhetorical graphs to model cross-passage coherence. These structures are jointly integrated into a planning blueprint that conditions the generation. Experiments on question answering and long-document summarization benchmarks show the efficacy of our approach. Disco-RAG achieves state-of-the-art results on the benchmarks without fine-tuning. These findings underscore the important role of discourse structure in advancing RAG systems.

Yuheng Chen, Teng Hu, Jiangning Zhang et al.

Recent advances in text-to-video diffusion models have enabled the generation of high-quality videos conditioned on textual descriptions. However, most existing text-to-video models rely solely on textual conditions, lacking general fine-grained controllability over video generation. To address this challenge, we propose InstanceV, a video generation framework that enables i) instance-level control and ii) global semantic consistency. Specifically, with the aid of proposed Instance-aware Masked Cross-Attention mechanism, InstanceV maximizes the utilization of additional instance-level grounding information to generate correctly attributed instances at designated spatial locations. To improve overall consistency, We introduce the Shared Timestep-Adaptive Prompt Enhancement module, which connects local instances with global semantics in a parameter-efficient manner. Furthermore, we incorporate Spatially-Aware Unconditional Guidance during both training and inference to alleviate the disappearance of small instances. Finally, we propose a new benchmark, named InstanceBench, which combines general video quality metrics with instance-aware metrics for more comprehensive evaluation on instance-level video generation. Extensive experiments demonstrate that InstanceV not only achieves remarkable instance-level controllability in video generation, but also outperforms existing state-of-the-art models in both general quality and instance-aware metrics across qualitative and quantitative evaluations.

Xurong Xie, Zhucun Xue, Jiafu Wu et al.

Knowledge distillation (KD) is a key technique for compressing large-scale language models (LLMs), yet prevailing logit-based methods typically employ static strategies that are misaligned with the dynamic learning process of student models. These methods typically treat all tokens indiscriminately and apply a single, fixed temperature, resulting in suboptimal knowledge transfer. To address these limitations, we propose LLM-Oriented Token-Adaptive Knowledge Distillation (AdaKD), a novel framework that adapts the distillation process to the real-time learning state of each token. AdaKD consists of two synergistic modules driven by a unified token difficulty metric. First, our Loss-Driven Adaptive Token Focusing (LATF) module dynamically adjusts the distillation focus by monitoring the student's learning stability, concentrating computational resources on the most valuable tokens at each training phase. Second, we introduce Inverse Difficulty Temperature Scaling (IDTS), a counterintuitive yet effective token-level temperature strategy. It employs low temperatures for difficult tokens for targeted error correction, and high temperatures for easy tokens to encourage students to learn from the teacher's complete and smooth output distribution, thereby enhancing generalization. As a plug-and-play framework, AdaKD can consistently improve the performance of various distillation methods on multiple model architectures and benchmarks.

Yijia Hong, Jiangning Zhang, Ran Yi et al.

Generating intermediate video content of varying lengths based on given first and last frames, along with text prompt information, offers significant research and application potential. However, traditional frame interpolation tasks primarily focus on scenarios with a small number of frames, no text control, and minimal differences between the first and last frames. Recent community developers have utilized large video models represented by Wan to endow frame-to-frame capabilities. However, these models can only generate a fixed number of frames and often fail to produce satisfactory results for certain frame lengths, while this setting lacks a clear official definition and a well-established benchmark. In this paper, we first propose a new practical Semantic Frame Interpolation (SFI) task from the perspective of academic definition, which covers the above two settings and supports inference at multiple frame rates. To achieve this goal, we propose a novel SemFi model building upon Wan2.1, which incorporates a Mixture-of-LoRA module to ensure the generation of high-consistency content that aligns with control conditions across various frame length limitations. Furthermore, we propose SFI-300K, the first general-purpose dataset and benchmark specifically designed for SFI. To support this, we collect and process data from the perspective of SFI, carefully designing evaluation metrics and methods to assess the model's performance across multiple dimensions, encompassing image and video, and various aspects, including consistency and diversity. Through extensive experiments on SFI-300K, we demonstrate that our method is particularly well-suited to meet the requirements of the SFI task.

Yuxuan Cai, Jiangning Zhang, Zhenye Gan et al.

Multimodal Large Language Models (MLLMs) have demonstrated significant advances in visual understanding tasks involving both images and videos. However, their capacity to comprehend human-centric video data remains underexplored, primarily due to the absence of comprehensive and high-quality evaluation benchmarks. Existing human-centric benchmarks predominantly emphasize video generation quality and action recognition, while overlooking essential perceptual and cognitive abilities required in human-centered scenarios. Furthermore, they are often limited by single-question paradigms and overly simplistic evaluation metrics. To address above limitations, we propose a modern HV-MMBench, a rigorously curated benchmark designed to provide a more holistic evaluation of MLLMs in human-centric video understanding. Compared to existing human-centric video benchmarks, our work offers the following key features: (1) Diverse evaluation dimensions: HV-MMBench encompasses 13 tasks, ranging from basic attribute perception (e.g., age estimation, emotion recognition) to advanced cognitive reasoning (e.g., social relationship prediction, intention prediction), enabling comprehensive assessment of model capabilities; (2) Varied data types: The benchmark includes multiple-choice, fill-in-blank, true/false, and open-ended question formats, combined with diverse evaluation metrics, to more accurately and robustly reflect model performance; (3) Multi-domain video coverage: The benchmark spans 50 distinct visual scenarios, enabling comprehensive evaluation across fine-grained scene variations; (4) Temporal coverage: The benchmark covers videos from short-term (10 seconds) to long-term (up to 30min) durations, supporting systematic analysis of models temporal reasoning abilities across diverse contextual lengths.

Jiangning Zhang, Liang Liu, Zhucun Xue et al.

Audio-guided face reenactment aims at generating photorealistic faces using audio information while maintaining the same facial movement as when speaking to a real person. However, existing methods can not generate vivid face images or only reenact low-resolution faces, which limits the application value. To solve those problems, we propose a novel deep neural network named APB2Face, which consists of GeometryPredictor and FaceReenactor modules. GeometryPredictor uses extra head pose and blink state signals as well as audio to predict the latent landmark geometry information, while FaceReenactor inputs the face landmark image to reenact the photorealistic face. A new dataset AnnVI collected from YouTube is presented to support the approach, and experimental results indicate the superiority of our method than state-of-the-arts, whether in authenticity or controllability.

Zhucun Xue, Nan Xue, Gui-Song Xia et al.

This paper presents a new deep-learning based method to simultaneously calibrate the intrinsic parameters of fisheye lens and rectify the distorted images. Assuming that the distorted lines generated by fisheye projection should be straight after rectification, we propose a novel deep neural network to impose explicit geometry constraints onto processes of the fisheye lens calibration and the distorted image rectification. In addition, considering the nonlinearity of distortion distribution in fisheye images, the proposed network fully exploits multi-scale perception to equalize the rectification effects on the whole image. To train and evaluate the proposed model, we also create a new largescale dataset labeled with corresponding distortion parameters and well-annotated distorted lines. Compared with the state-of-the-art methods, our model achieves the best published rectification quality and the most accurate estimation of distortion parameters on a large set of synthetic and real fisheye images.