Searcharxiv⌕ Search

arXiv subjects

Yunhang Shen

Publications and source records attributed to Yunhang Shen.

At least 19 recordsLinked to original sources
arXiv2026

Video-IFBench: Evaluating Instruction Following of Multimodal LLMs in Video Understanding Scenarios

Multimodal Large Language Models (MLLMs) have shown strong performance in video understanding. However, their ability to follow instructions in this domain remains under-explored. Real-world video understanding requires models not only to interpret video content correctly, but also to satisfy diverse user-specified constraints. Existing benchmarks focus primarily on task accuracy rather than instruction adherence, leaving this capability insufficiently evaluated. To address this gap, we introduce Video-IFBench, a comprehensive benchmark for evaluating instruction following in video understanding, where models must satisfy diverse user-specified constraints, including those grounded in visual and audio content. We develop an instruction taxonomy with four templates, including single-task, multi-task, selection, and nested instructions, covering 32 task types and 39 manually designed constraint categories spanning both semantic and format requirements. To reduce annotation cost, we build a semi-automatic data construction pipeline that combines MLLMs, programmatic processing, and human verification, resulting in 1.5K samples. We conduct a large-scale evaluation of more than 20 recent MLLMs and show that video instruction following remains challenging for current models, especially for instructions with many constraints, semantic constraints, or complex conditional structures that require selecting the correct branch or path based on video content. We hope our work will facilitate future research on instruction following in video understanding scenarios.

cs.CV↗
arXiv2026

FIRM-Video: Check Before You Score for Reliable Text-to-Video Reward Modeling

Reliable reward models are essential for text-to-video evaluation and alignment. However, the trade-off between evaluation accuracy and inference efficiency places high demands on the quality of training supervision. Existing approaches often rely on holistic judges with fixed rubrics or open-ended reasoning, leading to incomplete inspection, unfaithful justification, and entangled attribution. We introduce FIRM-Video, a unified checklist-driven data construction framework based on a check-before-score principle: construct dimension-specific checklists, verify each criterion against temporal visual evidence, and aggregate only verified decisions. For Instruction Following, FIRM-Video decomposes prompts into weighted atomic requirements; for World Coherence, it constructs prompt-calibrated, target-specific checks grounded in visible entities and actions; and for Perceptual Quality, it applies a generic taxonomy of visual defects. The verified criteria and scores are further transformed into natural-language analyses for end-to-end reward modeling. Subsequently, we construct FIRM-Video-90K with 88,044 dimension-specific instances from 29,348 videos, and introduce FIRM-Video-Bench with 750 point-wise human annotations across 250 videos. The Qwen3-VL-based FIRM-Video-8B achieves the best overall MAE on FIRM-Video-Bench while consistently delivering the highest VBench Total, Quality, and Semantic Scores in Best-of-8 sampling across three video generators.

cs.CV↗
arXiv2026

Diagnosing and Mitigating Perception-Decision Misalignment in Omni-LLMs via Modality Subspace Activation

Omni-Large Language Models (Omni-LLMs) power complex multi-modal reasoning in applications like World Action Models and autonomous agents. However, their strong performance often masks a profound Perceptual-Decision Misalignment (PDM), where decisions remain unfaithful to multi-modal perceptions. To diagnose this, we formalize Causal Modality Sensitivity (CMS), operationalized via a dual-lens framework: Answer Retention Rate (ARR) at the macro behavioral level, and Logit Angular Discrepancy (LAD) to track microscopic distribution shifts. We also curate CausalMSBench, a diagnostic dataset isolating language priors. Benchmarking reveals that popular Omni-LLMs exhibit critically low CMS, showing negligible distribution shifts even when key modalities are removed. To rectify this, we propose Modality Subspace Activation (MSA), a training-free inference-time framework that uses Singular Value Decomposition (SVD) to estimate modal activation strengths. MSA dynamically balances modal projections in the last hidden state, effectively restoring CMS across benchmarks.

cs.LG↗
arXiv2026

Omni-Diffusion: Unified Multimodal Understanding and Generation with Masked Discrete Diffusion

While recent multimodal large language models (MLLMs) have made impressive strides, they predominantly employ a conventional autoregressive architecture as their backbone, leaving significant room to explore effective and efficient alternatives in architectural design. Concurrently, recent studies have successfully applied discrete diffusion models to various domains, such as visual understanding and image generation, revealing their considerable potential as a promising backbone for multimodal systems. Drawing inspiration from these pioneering studies, we introduce Omni-Diffusion, the first any-to-any multimodal language model built entirely on mask-based discrete diffusion models, which unifies understanding and generation across text, speech, and images. Omni-Diffusion employs a unified mask-based discrete diffusion model to directly capture the joint distribution over discrete multimodal tokens. This approach supports not only bimodal tasks but also more complex scenarios involving multiple modalities. On a diverse set of benchmarks, our method outperforms or performs on par with existing multimodal systems that process two or more modalities, highlighting the significant promise of diffusion models in powering the next generation of multimodal foundation models. Project webpage: https://omni-diffusion.github.io.

cs.CV↗
arXiv2026

Towards Fast and Effective Long Video Understanding of Multimodal Large Language Models via Adaptive Quasi-Gaussian Sampling

Long video understanding remains a daunting challenge for Multimodal Large Language Models (MLLMs) due to the excessive computation and memory footprint. Thus, keyframe selection is often adopted to mitigate this shortcoming, which however still suffers from low flexibility and high noise due to its hard sampling principle. In this paper, we define video frame selection as a problem of Quasi-Gaussian Sampling, and propose an adaptive and training-free approach termed AdaQ. Inspired by the 3-$σ$ rule of Gaussian distribution, the objective of AdaQ is to achieve the optimal 3-$σ$ interval for different examples, i.e., a smaller 3-$σ$ interval for the local query and a larger one for the global query, thereby facilitating robust and adaptive frame sampling. To validate AdaQ, we apply it to four MLLMs with three embedding models. The extensive experimental results not only show its obvious performance gains over the default MLLMs and the SOTA keyframe selection methods, e.g., helping Qwen3-VL-8B outperform GPT4o by 15.8% on average by using only 64 frames, but also confirm its superior robustness and high efficiency for long-video understanding, e.g., only 1 hyper-parameter needs to be set.

cs.CV↗
arXiv2026

SMART: Shot-Aware Multimodal Video Moment Retrieval with Audio-Enhanced MLLM

Video Moment Retrieval is a task in video understanding that aims to localize a specific temporal segment in an untrimmed video based on a natural language query. Despite recent progress in moment retrieval from videos using both traditional techniques and Multimodal Large Language Models (MLLM), most existing methods still rely on coarse temporal understanding and a single visual modality, limiting performance on complex videos. To address this, we introduce \textit{S}hot-aware \textit{M}ultimodal \textit{A}udio-enhanced \textit{R}etrieval of \textit{T}emporal \textit{S}egments (SMART), an MLLM-based framework that integrates audio cues and leverages shot-level temporal structure. SMART enriches multimodal representations by combining audio and visual features while applying \textbf{Shot-aware Token Compression}, which selectively retains high-information tokens within each shot to reduce redundancy and preserve fine-grained temporal details. We also refine prompt design to better utilize audio-visual cues. Evaluations on Charades-STA and QVHighlights show that SMART achieves significant improvements over state-of-the-art methods, including a 1.61\% increase in R1@0.5 and 2.59\% gain in R1@0.7 on Charades-STA.

cs.CV↗
arXiv2026

Toward Native Multimodal Modeling: A Roadmap

Multimodal modeling represents a vital step from modality-agnostic reasoning toward world modeling. While early approaches predominantly rely on late-fusion that assembles encoders and frozen language backbones with output heads, recent efforts have shifted the paradigm toward native multimodal modeling (NMM) with the intrinsic integration of modalities for superior multimodal performance. Despite its potential, the design space of native architectures remains insufficiently defined. In this paper, we present the community with a formalized roadmap for this transition. Specifically, we formally define the architectural nativity, distinguishing mid-fusion and early-fusion from non-native paradigms. We further organize the existing native models through the lens of input-output duality into three categories: (i) Multi-to-Text for cross-modal comprehension with text-only output; (ii) Multi-to-Target for scenario-oriented generation, e.g., image, audio and video generation, and (iii) Multi-to-Multi for unified modeling with symmetric input-output. We deliver a comprehensive and industrial-grade investigation into the transition toward the definitive NMM framework, where understanding and generation seamlessly coexist within a unified transformer paradigm. We systematically unpack the end-to-end pipeline from industrial perspectives from architectural coordination, massive data curation, to full-stack training recipes, inference & deployment, and the comprehensive evaluation for truly native modeling.

cs.CV↗
arXiv2026

GroupGPT: A Token-efficient and Privacy-preserving Agentic Framework for Multi-User Chat Assistant

Recent advances in large language models (LLMs) have enabled increasingly capable chatbots. However, most existing systems focus on single-user settings and do not generalize well to multi-user group chat interactions, where agents require more proactive and accurate intervention under complex, evolving contexts. Existing approaches typically rely on LLMs for both intervention reasoning and response generation, leading to high token consumption, limited scalability, and potential privacy risks. To address these challenges, we propose GroupGPT, a token-efficient and privacy-preserving agentic framework for multi-user chat assistant. GroupGPT adopts an edge-cloud model collaboration architecture to decouple intervention timing from response generation, enabling efficient and accurate decision-making while preserving user privacy through on-device processing of sensitive information. The framework also supports multimodal inputs, including memes, images, videos, and voice messages.To support evaluation of timing accuracy and response quality, we further introduce MUIR, a benchmark dataset for multi-user chat assistant intervention reasoning. MUIR contains 2,500 annotated group chat segments with intervention labels and rationales. We evaluate a range of models on MUIR, spanning from open-source to proprietary variants, including both LLMs and their smaller counterparts. Extensive experiments demonstrate that GroupGPT generates accurate and well-timed responses, achieving an average score of 4.72/5.0 in LLM-based evaluation, and is well-received by users across diverse group chat scenarios. Moreover, GroupGPT reduces the token usage by up to 3 times compared to baselines, while providing privacy sanitization of user messages before cloud transmission. Code is available at: https://github.com/Eliot-Shen/GroupGPT .

cs.CL↗
arXiv2026

Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understanding

With the rapid advancement of video understanding, existing benchmarks are becoming increasingly saturated, exposing a critical discrepancy between inflated leaderboard scores and real-world model capabilities. To address this widening gap, we introduce Video-MME-v2, a comprehensive benchmark designed to rigorously evaluate the robustness and faithfulness of video understanding. To systematically evaluate model capabilities, we design a \textbf{progressive tri-level hierarchy} that incrementally increases the complexity of video comprehension, ranging from multi-point visual information aggregation, to temporal dynamics modeling, and ultimately to complex multimodal reasoning. Besides, in contrast to conventional per-question accuracy, we propose a \textbf{group-based non-linear evaluation} strategy that enforces both consistency across related queries and coherence in multi-step reasoning. It penalizes fragmented or guess-based correctness and assigns credit only to answers supported by valid reasoning. To guarantee data quality, Video-MME-v2 is constructed through a rigorously controlled human annotation pipeline, involving 12 annotators and 50 independent reviewers. Backed by \textbf{3,300 human-hours} and up to \textbf{5 rounds} of quality assurance, Video-MME-v2 aims to serve as one of the most authoritative video benchmarks. Extensive experiments reveal a substantial gap between current best model Gemini-3-Pro and human experts, and uncover a clear hierarchical bottleneck where errors in visual information aggregation and temporal modeling propagate to limit high-level reasoning. We further find that thinking-based reasoning is highly dependent on textual cues, improving performance with subtitles but sometimes degrading it in purely visual settings. By exposing these limitations, Video-MME-v2 establishes a demanding new testbed for the development of next-generation video MLLMs.

cs.CV↗
arXiv2026

MASA: Rethinking the Representational Bottleneck in LoRA with Multi-A Shared Adaptation

Low-Rank Adaptation (LoRA) has emerged as a dominant method in Parameter-Efficient Fine-Tuning (PEFT) for large language models, which augments the transformer layer with one down-projection $A$ and one up-projection $B$. However, LoRA's reliance on a single down-projection matrix ($A$) creates a representational bottleneck, as this solitary feature extractor is inherently insufficient for capturing the diverse signals required by complex tasks. This motivates our architectural shift to focus on enriching the feature adaptation to improve the downstream task adaptation ability. We propose MASA (Multi-$A$ Shared Adaptation), an architecture that implements a multi-$A$, single-$B$ structure where the multi-$A$ expert ensemble is asymmetrically shared across layers to ensure parameter efficiency. In MASA, these specialized experts capture diverse features, which are then integrated by a single, layer-specific $B$-matrix. The effectiveness and versatility of our method are validated through a comprehensive suite of experiments spanning multi-domain generalization, single-domain specialization, and multi-task reasoning. For example, on the MMLU benchmark, MASA achieves an average accuracy of 59.62%, outperforming the standard LoRA by 1.08 points (a relative improvement of 1.84%) with comparable learnable parameters of 0.52%.

cs.CL↗
arXiv2026

Can Unified Generation and Understanding Models Maintain Semantic Equivalence Across Different Output Modalities?

Unified Multimodal Large Language Models (U-MLLMs) integrate understanding and generation within a single architecture. However, existing evaluations typically assess these capabilities separately, overlooking semantic equivalence, i.e., the ability to manifest consistent reasoning results regardless of the output modality. In this work, we investigate whether current U-MLLMs satisfy this premise. We observe that while models demonstrate robust textual reasoning, they fail to maintain semantic equivalence when required to render the same results in the image modality. To rigorously diagnose this discrepancy, we introduce VGUBench, a framework to decouple reasoning logic from generation fidelity. VGUBench comprises three diagnostic tasks: (1)Textual Generative Understanding, establishing a baseline for reasoning accuracy in textual response; (2)Visual Generative Understanding, evaluating the ability to generate visual responses that represent the correct answer; and (3)a Visual Rendering control task, which assesses the ability to directly render explicit visual descriptions into images without complex reasoning. Our evaluation reveals a significant disparity: despite strong performance in textual understanding and visual rendering, U-MLLMs exhibit a marked performance collapse when required to generate visual answers to questions. Furthermore, we find a negligible correlation between visual answering performance and basic rendering quality. These results suggest that the failure stems not from insufficient generation fidelity, but from a breakdown in cross-modal semantic alignment. We provide diagnostic insights to address this challenge in future Unified Generation and Understanding Models.

cs.CV↗
arXiv2026

Unleashing MLLMs on the Edge: A Unified Framework for Cross-Modal ReID via Adaptive SVD Distillation

Practical cloud-edge deployment of Cross-Modal Re-identification (CM-ReID) faces challenges due to maintaining a fragmented ecosystem of specialized cloud models for diverse modalities. While Multi-Modal Large Language Models (MLLMs) offer strong unification potential, existing approaches fail to adapt them into a single end-to-end backbone and lack effective knowledge distillation strategies for edge deployment. To address these limitations, we propose MLLMEmbed-ReID, a unified framework based on a powerful cloud-edge architecture. First, we adapt a foundational MLLM into a state-of-the-art cloud model. We leverage instruction-based prompting to guide the MLLM in generating a unified embedding space across RGB, infrared, sketch, and text modalities. This model is then trained efficiently with a hierarchical Low-Rank Adaptation finetuning (LoRA-SFT) strategy, optimized under a holistic cross-modal alignment objective. Second, to deploy its knowledge onto an edge-native student, we introduce a novel distillation strategy motivated by the low-rank property in the teacher's feature space. To prioritize essential information, this method employs a Principal Component Mapping loss, while relational structures are preserved via a Feature Relation loss. Our lightweight edge-based model achieves state-of-the-art performance on multiple visual CM-ReID benchmarks, while its cloud-based counterpart excels across all CM-ReID benchmarks. The MLLMEmbed-ReID framework thus presents a complete and effective solution for deploying unified MLLM-level intelligence on resource-constrained devices. The code and models will be open-sourced soon.

cs.CV↗
arXiv2026

SE-Search: Self-Evolving Search Agent via Memory and Dense Reward

Retrieval augmented generation (RAG) reduces hallucinations and factual errors in large language models (LLMs) by conditioning generation on retrieved external knowledge. Recent search agents further cast RAG as an autonomous, multi-turn information-seeking process. However, existing methods often accumulate irrelevant or noisy documents and rely on sparse reinforcement learning signals. We propose \textbf{S}elf-\textbf{E}volving \textbf{Search}, a Self-Evolving Search agent that improves online search behavior through three components, memory purification, atomic query training, and dense rewards. SE-Search follows a \textit{Think-Search-Memorize} strategy that retains salient evidence while filtering irrelevant content. Atomic query training promotes shorter and more diverse queries, improving evidence acquisition. Dense rewards provide fine-grained feedback that speeds training. Experiments on single-hop and multi-hop question answering benchmarks show that \texttt{SE-Search-3B} outperforms strong baselines, yielding a $10.8$ point absolute improvement and a $33.8\%$ relative gain over Search-R1.\footnote{We will make the code and model weights publicly available upon acceptance.}

cs.CL↗
arXiv2026

Youtu-VL: Unleashing Visual Potential via Unified Vision-Language Supervision

Despite the significant advancements represented by Vision-Language Models (VLMs), current architectures often exhibit limitations in retaining fine-grained visual information, leading to coarse-grained multimodal comprehension. We attribute this deficiency to a suboptimal training paradigm inherent in prevailing VLMs, which exhibits a text-dominant optimization bias by conceptualizing visual signals merely as passive conditional inputs rather than supervisory targets. To mitigate this, we introduce Youtu-VL, a framework leveraging the Vision-Language Unified Autoregressive Supervision (VLUAS) paradigm, which fundamentally shifts the optimization objective from ``vision-as-input'' to ``vision-as-target.'' By integrating visual tokens directly into the prediction stream, Youtu-VL applies unified autoregressive supervision to both visual details and linguistic content. Furthermore, we extend this paradigm to encompass vision-centric tasks, enabling a standard VLM to perform vision-centric tasks without task-specific additions. Extensive empirical evaluations demonstrate that Youtu-VL achieves competitive performance on both general multimodal tasks and vision-centric tasks, establishing a robust foundation for the development of comprehensive generalist visual agents.

cs.CV↗
arXiv2025

Omni-Referring Image Segmentation

In this paper, we propose a novel task termed Omni-Referring Image Segmentation (OmniRIS) towards highly generalized image segmentation. Compared with existing unimodally conditioned segmentation tasks, such as RIS and visual RIS, OmniRIS supports the input of text instructions and reference images with masks, boxes or scribbles as omni-prompts. This property makes it can well exploit the intrinsic merits of both text and visual modalities, i.e., granular attribute referring and uncommon object grounding, respectively. Besides, OmniRIS can also handle various segmentation settings, such as one v.s. many and many v.s. many, further facilitating its practical use. To promote the research of OmniRIS, we also rigorously design and construct a large dataset termed OmniRef, which consists of 186,939 omni-prompts for 30,956 images, and establish a comprehensive evaluation system. Moreover, a strong and general baseline termed OmniSegNet is also proposed to tackle the key challenges of OmniRIS, such as omni-prompt encoding. The extensive experiments not only validate the capability of OmniSegNet in following omni-modal instructions, but also show the superiority of OmniRIS for highly generalized image segmentation.

cs.CV↗
arXiv2025

Long-VITA: Scaling Large Multi-modal Models to 1 Million Tokens with Leading Short-Context Accuracy

We introduce Long-VITA, a simple yet effective large multi-modal model for long-context visual-language understanding tasks. It is adept at concurrently processing and analyzing modalities of image, video, and text over 4K frames or 1M tokens while delivering advanced performances on short-context multi-modal tasks. We propose an effective multi-modal training schema that starts with large language models and proceeds through vision-language alignment, general knowledge learning, and two sequential stages of long-sequence fine-tuning. We further implement context-parallelism distributed inference and logits-masked language modeling head to scale Long-VITA to infinitely long inputs of images and texts during model inference. Regarding training data, Long-VITA is built on a mix of 17M samples from public datasets only and demonstrates state-of-the-art performance on various multi-modal benchmarks, compared against recent cutting-edge models with internal data. Long-VITA is fully open-source and reproducible.. By leveraging our inference designs, Long-VITA models achieve a remarkable 2x prefill speedup and 4x context length extension in a single node with 8 GPUs. We hope Long-VITA can serve as a competitive baseline and offer valuable insights for the open-source community in advancing long-context multi-modal understanding.

cs.CV↗
arXiv2025

DeepOmni: Towards Seamless and Smart Speech Interaction with Adaptive Modality-Specific MoE

Native multimodal large language models (MLLMs) restructure a single large language model (LLM) into a spoken language model (SLM) capable of both speech and text generation. Compared to modular and aligned MLLMs, native MLLMs preserve richer paralinguistic features such as emotion and prosody, and generate speech responses directly within the backbone LLM rather than using a separate speech decoder. This integration also results in lower response latency and smoother interaction. However, native MLLMs suffer from catastrophic forgetting and performance degradation because the available paired speech-text data is insufficient to support the pretraining of MLLMs compared to the vast amount of text data required to pretrain text LLMs. To address this issue, we propose DeepTalk, a framework for adaptive modality expert learning based on a Mixture of Experts (MoE) architecture. DeepTalk first adaptively distinguishes modality experts according to their modality load within the LLM. Each modality expert then undergoes specialized single-modality training, followed by joint multimodal collaborative training. As a result, DeepTalk incurs only a 5.5% performance drop compared to the original LLM, which is significantly lower than the average performance drop of over 20% typically seen in native MLLMs (such as GLM-4-Voice), and is on par with modular MLLMs. Meanwhile, the end-to-end dialogue latency remains within 0.5 seconds, ensuring a seamless and intelligent speech interaction experience. Code and models are released at https://github.com/talkking/DeepTalk.

cs.CL↗
arXiv2025

MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models

Multimodal Large Language Model (MLLM) relies on the powerful LLM to perform multimodal tasks, showing amazing emergent abilities in recent studies, such as writing poems based on an image. However, it is difficult for these case studies to fully reflect the performance of MLLM, lacking a comprehensive evaluation. In this paper, we fill in this blank, presenting the first comprehensive MLLM Evaluation benchmark MME. It measures both perception and cognition abilities on a total of 14 subtasks. In order to avoid data leakage that may arise from direct use of public datasets for evaluation, the annotations of instruction-answer pairs are all manually designed. The concise instruction design allows us to fairly compare MLLMs, instead of struggling in prompt engineering. Besides, with such an instruction, we can also easily carry out quantitative statistics. A total of 30 advanced MLLMs are comprehensively evaluated on our MME, which not only suggests that existing MLLMs still have a large room for improvement, but also reveals the potential directions for the subsequent model optimization. The data are released at the project page https://github.com/BradyFU/Awesome-Multimodal-Large-Language-Models/tree/Evaluation.

cs.CV↗