Multimodal Large Language Models

Also known as MLLM

Ask about this topic

Overview

Multimodal large language models connect language generation and reasoning to visual, audio, video, and sometimes sensor or structured data. Common approaches align pretrained modality encoders with a language model using projectors or adapters, or train models to process modalities jointly; instruction tuning, reinforcement learning, retrieval, and tool use extend their capabilities. Evaluation spans visual question answering, document reading, long-video and spatial reasoning, and specialized settings such as medical imaging, where models still often miss fine-grained evidence or fail to use it consistently. A central tension is how to improve cross-modal grounding, temporal and physical reasoning, and safety while reducing the high inference costs of processing long or detailed inputs.

Latest papers

CardsList

1,096 papers106 papers in the last four weeks, up 23% on the four weeks before. 0.6% of all new papers.

Mar 31, 2026

Xuanwu: Evolving General Multimodal Models into an Industrial-Grade Foundation for Content Ecosystems

In recent years, multimodal large models have continued to improve on general benchmarks. However, in real-world content moderation and adversarial settings, mainstream models still suffer from degraded generalization and catastrophic forgetting because of limited fine-grained visual perception and insufficient modeling of long-tail noise. In this paper, we present Xuanwu VL-2B as a case study of how general multimodal models can be developed into an industrial-grade foundation model for content ecosystems. The model adopts a compact InternViT-300M + MLP + Qwen3 1.7B architecture, balancing fine-grained visual perception, language-semantic alignment, and deployment cost within an approximately 2B-parameter budget. To balance business specialization with the retention of general capabilities, we developed a data iteration and curation mechanism and trained the model through a progressive three-stage pipeline: pre-training, mid-training, and post-training. Ablation studies and offline business evaluations show that Xuanwu VL-2B achieves an average score of 67.90 across seven OpenCompass multimodal metrics (vs. 64.27 for InternVL 3.5 2B), an average recall of 94.38% over seven independent business moderation tasks, and a weighted overall recall of 82.82% on policy-violating text in challenging adversarial OCR scenarios, outperforming Gemini-2.5-Pro (76.72%). These results show that, under a limited parameter budget, Xuanwu VL-2B achieves a practical balance among business alignment, visual perception, general capability retention, and deployment cost.

Zhiqian Zhang, Xu Zhao, Xiaoqing Xu +5

Computational Intelligence Dept, Hello Group Inc.

Multimodal Content ModerationMultimodal Large Language ModelsMultimodal RobustnessMultimodal Pretraining
AskPDF
Mar 30, 2026

AdaptToken: Entropy-based Adaptive Token Selection for MLLM Long Video Understanding

Long video understanding remains challenging for Multi-modal Large Language Models (MLLMs) due to high memory costs and context-length limits. Prior approaches mitigate this by scoring and selecting frames/tokens within short clips, but they lack a principled mechanism to (i) compare relevance across distant video clips and (ii) stop processing once sufficient evidence has been gathered. We propose AdaptToken, a training-free framework that turns an MLLM's self-uncertainty into a global control signal for long-video token selection. AdaptToken splits a video into groups, extracts cross-modal attention to rank tokens within each group, and uses the model's response entropy to estimate each group's prompt relevance. This entropy signal enables a global token budget allocation across groups and further supports early stopping (AdaptToken-Lite), skipping the remaining groups when the model becomes sufficiently certain. Across four long-video benchmarks (VideoMME, LongVideoBench, LVBench, and MLVU) and multiple base MLLMs (7B-72B), AdaptToken consistently improves accuracy (e.g., +6.7 on average over Qwen2.5-VL 7B) and continues to benefit from extremely long inputs (up to 10K frames), while AdaptToken-Lite reduces inference time by about half with comparable performance. Project page: https://haozheqi.github.io/adapt-token

Haozhe Qi, Kevin Qu, Mahdi Rad +3

Microsoft Spatial AI Lab · EPFL · ETH Zurich

Long-Video UnderstandingMultimodal Large Language ModelsLLM Inference Acceleration
AskPDF
Mar 30, 2026

Seeing with You: Perception-Reasoning Coevolution for Multimodal Reasoning

Reinforcement learning with verifiable rewards (RLVR) has substantially enhanced the reasoning capabilities of multimodal large language models (MLLMs). However, existing RLVR approaches typically rely on outcome-driven optimization that updates both perception and reasoning using a shared reward based solely on the final answer. This shared reward blurs credit assignment, frequently improving reasoning patterns while failing to reliably enhance the accuracy of upstream visual evidence extraction. To address this perception bottleneck, we introduce PRCO (Perception-Reasoning Coevolution), a dual-role RLVR framework with a shared policy. PRCO consists of two cooperative roles: an Observer that generates an evidence caption tailored to the question and a Solver that predicts the final answer based on this caption. Crucially, PRCO employs role-specific reward signals: the Solver is optimized using verifiable outcome rewards on the final answer, while the Observer receives a utility reward derived from the Solver's downstream success. Extensive experiments across eight challenging multimodal reasoning benchmarks demonstrate that PRCO yields consistent improvements across model scales by over 7 points on average accuracy compared to the base model, outperforming prior open-source RL-tuned baselines.

Ziqi Miao, Haonan Jia, Lijun Li +4

Multimodal ReasoningMultimodal Large Language ModelsRL for VLMsReinforcement Learning with Verifiable Rewards
AskPDF
Mar 30, 2026

ResAdapt: Adaptive Resolution for Efficient Multimodal Reasoning

Multimodal Large Language Models (MLLMs) achieve stronger visual understanding by scaling input fidelity, yet the resulting visual token growth makes jointly sustaining high spatial resolution and long temporal context prohibitive. We argue that the bottleneck lies not in how post-encoding representations are compressed but in the volume of pixels the encoder receives, and address it with ResAdapt, an Input-side adaptation framework that learns how much visual budget each frame should receive before encoding. ResAdapt couples a lightweight Allocator with an unchanged MLLM backbone, so the backbone retains its native visual-token interface while receiving an operator-transformed input. We formulate allocation as a contextual bandit and train the Allocator with Cost-Aware Policy Optimization (CAPO), which converts sparse rollout feedback into a stable accuracy-cost learning signal. Across budget-controlled video QA, temporal grounding, and image reasoning tasks, ResAdapt improves low-budget operating points and often lies on or near the efficiency-accuracy frontier, with the clearest gains on reasoning-intensive benchmarks under aggressive compression. Notably, ResAdapt supports up to 16x more frames at the same visual budget while delivering over 15% performance gain. Code is available at https://github.com/Xnhyacinth/ResAdapt.

Huanxuan Liao, Zhongtao Jiang, Yupu Hao +6

Institute of Automation, Chinese Academy of Sciences · University of Chinese Academy of Sciences

Efficient Multimodal InferenceMultimodal Large Language Models
AskPDF
Mar 30, 2026

AutoCut: End-to-end advertisement video editing based on multimodal discretization and controllable generation

Short-form videos have become a primary medium for digital advertising, requiring scalable and efficient content creation. However, current workflows and AI tools remain disjoint and modality-specific, leading to high production costs and low overall efficiency. To address this issue, we propose AutoCut, an end-to-end advertisement video editing framework based on multimodal discretization and controllable editing. AutoCut employs dedicated encoders to extract video and audio features, then applies residual vector quantization to discretize them into unified tokens aligned with textual representations, constructing a shared video-audio-text token space. Built upon a foundation model, we further develop a multimodal large language model for video editing through combined multimodal alignment and supervised fine-tuning, supporting tasks covering video selection and ordering, script generation, and background music selection within a unified editing framework. Finally, a complete production pipeline converts the predicted token sequences into deployable long video outputs. Experiments on real-world advertisement datasets show that AutoCut reduces production cost and iteration time while substantially improving consistency and controllability, paving the way for scalable video creation.

Milton Zhou, Sizhong Qin, Yongzhi Li +2

Tsinghua University · Kuaishou Technology

Video EditingMultimodal Large Language Models
AskPDF
Mar 30, 2026

Beyond Static Vision: Scene Dynamic Field Unlocks Intuitive Physics Understanding in Multi-modal Large Language Models

While Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities in image and video understanding, their ability to comprehend the physical world has become an increasingly important research focus. Despite their improvements, current MLLMs struggle significantly with high-level physics reasoning. In this work, we investigate the first step of physical reasoning, i.e., intuitive physics understanding, revealing substantial limitations in understanding the dynamics of continuum objects. To isolate and evaluate this specific capability, we introduce two fundamental benchmark tasks: Next Frame Selection (NFS) and Temporal Coherence Verification (TCV). Our experiments demonstrate that even state-of-the-art MLLMs perform poorly on these foundational tasks. To address this limitation, we propose Scene Dynamic Field (SDF), a concise approach that leverages physics simulators within a multi-task fine-tuning framework. SDF substantially improves performance, achieving up to 20.7% gains on fluid tasks while showing strong generalization to unseen physical domains. This work not only highlights a critical gap in current MLLMs but also presents a promising cost-efficient approach for developing more physically grounded MLLMs. Our code and data are available at https://github.com/andylinx/Scene-Dynamic-Field.

Nanxi Li, Xiang Wang, Yuanjie Chen +3

Shanghai Jiao Tong University · Tianjin University · Shanghai Innovation Institute

Physical ReasoningMultimodal Large Language Models
AskPDF
Mar 30, 2026

CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs

Analogical reasoning tests a fundamental aspect of human cognition: mapping the relation from one pair of objects to another. Existing evaluations of this ability in multimodal large language models (MLLMs) overlook the ability to compose rules from multiple sources, a critical component of higher-order intelligence. To close this gap, we introduce CARV (Compositional Analogical Reasoning in Vision), a novel task together with a 5,500-sample dataset as the first diagnostic benchmark. We extend the analogy from a single pair to multiple pairs, which requires MLLMs to extract symbolic rules from each pair and compose new transformations. Evaluation on the state-of-the-art MLLMs reveals a striking performance gap: even Gemini-2.5 Pro achieving only 40.4% accuracy, far below human-level performance of 100%. Diagnostic analysis shows two consistent failure modes: (1) decomposing visual changes into symbolic rules, and (2) maintaining robustness under diverse or complex settings, highlighting the limitations of current MLLMs on this task.

Yongkang Du, Xiaohan Zou, Minhao Cheng +1

Pennsylvania State University

Analogical ReasoningMultimodal ReasoningMultimodal Large Language ModelsMultimodal Model Evaluation
AskPDF
Mar 30, 2026

Adversarial Attacks on Multimodal Large Language Models: A Comprehensive Survey

Multimodal large language models (MLLMs) integrate information from multiple modalities such as text, images, audio, and video, enabling complex capabilities such as visual question answering and audio translation. While powerful, this increased expressiveness introduces new and amplified vulnerabilities to adversarial manipulation. This survey provides a comprehensive and systematic analysis of adversarial threats to MLLMs, moving beyond enumerating attack techniques to explain the underlying causes of model susceptibility. We introduce a taxonomy that organizes adversarial attacks according to attacker objectives, unifying diverse attack surfaces across modalities and deployment settings. Additionally, we also present a vulnerability-centric analysis that links integrity attacks, safety and jailbreak failures, control and instruction hijacking, and training-time poisoning to shared architectural and representational weaknesses in multimodal systems. Together, this framework provides an explanatory foundation for understanding adversarial behavior in MLLMs and informs the development of more robust and secure multimodal language systems.

Bhavuk Jain, Sercan Ö. Arık, Hardeo K. Thakur

Google · Bennett University, India

Adversarial Attacks on Unified Multimodal ModelsLLM SecurityMultimodal RobustnessMultimodal Large Language Models
AskPDF
Mar 29, 2026

Learning to See through Illumination Extremes with Event Streaming in Multimodal Large Language Models

Multimodal Large Language Models (MLLMs) perform strong vision-language reasoning under standard conditions but fail in extreme illumination, where RGB inputs lose irrevocable structure and semantics. We propose Event-MLLM, an event-enhanced model that performs all-light visual reasoning by dynamically fusing event streams with RGB frames. Two key components drive our approach: an Illumination Indicator - a learnable signal derived from a DINOv2 branch that represents exposure degradation and adaptively modulates event-RGB fusion - and an Illumination Correction Loss that aligns fused features with non-degraded (normal-light) semantics in the latent space, compensating for information lost in extreme lighting. We curate the first multi-illumination event-instruction corpus for MLLMs, with 2,241 event-RGB samples (around 6 QA pairs each) across diverse scenes and 17 brightness rates (0.05x - 20x), plus an instruct-following benchmark for reasoning, counting, and fine-grained recognition under extreme lighting. Experiments show that Event-MLLM markedly outperforms general-purpose, illumination-adaptive, and event-only baselines, setting a new state of the art in robust multimodal perception and reasoning under challenging illumination.

Baoheng Zhang, Jiahui Liu, Gui Zhao +7

The University of Hong Kong

Vision-Language ModelsMultimodal Large Language ModelsVLM Robustness
AskPDF
Mar 27, 2026

RealBirdID: Benchmarking Bird Species Identification in the Era of MLLMs

Fine-grained bird species identification in the wild is frequently unanswerable from a single image: key cues may be non-visual (e.g. vocalization), or obscured due to occlusion, camera angle, or low resolution. Yet today's multimodal systems are typically judged on answerable, in-schema cases, encouraging confident guesses rather than principled abstention. We propose the RealBirdID benchmark: given an image of a bird, a system should either answer with a species or abstain with a concrete, evidence-based rationale: "requires vocalization," "low quality image," or "view obstructed". For each genus, the dataset includes a validation split composed of curated unanswerable examples with labeled rationales, paired with a companion set of clearly answerable instances. We find that (1) the species identification on the answerable set is challenging for a variety of open-source and proprietary models (less than 13% accuracy for MLLMs including GPT-5 and Gemini-2.5 Pro), (2) models with greater classification ability are not necessarily more calibrated to abstain from unanswerable examples, and (3) that MLLMs generally fail at providing correct reasons even when they do abstain. RealBirdID establishes a focused target for abstention-aware fine-grained recognition and a recipe for measuring progress.

Logan Lawrence, Mustafa Chasmai, Rangel Daroya +8

Computer Vision Lab, UMass Amherst

Fine-Grained Image ClassificationMultimodal Model EvaluationSelective PredictionMultimodal Large Language Models
AskPDF
Mar 27, 2026

MA-Bench: Towards Fine-grained Micro-Action Understanding

With the rapid development of Multimodal Large Language Models (MLLMs), their potential in Micro-Action understanding, a vital role in human emotion analysis, remains unexplored due to the absence of specialized benchmarks. To tackle this issue, we present MA-Bench, a benchmark comprising 1,000 videos and a three-tier evaluation architecture that progressively examines micro-action perception, relational comprehension, and interpretive reasoning. MA-Bench contains 12,000 structured question-answer pairs, enabling systematic assessment of both recognition accuracy and action interpretation. The results of 23 representative MLLMs reveal that there are significant challenges in capturing motion granularity and fine-grained body-part dynamics. To address these challenges, we further construct MA-Bench-Train, a large-scale training corpus with 20.5K videos annotated with structured micro-action captions for fine-tuning MLLMs. The results of Qwen3-VL-8B fine-tuned on MA-Bench-Train show clear performance improvements across micro-action reasoning and explanation tasks. Our work aims to establish a foundation benchmark for advancing MLLMs in understanding subtle micro-action and human-related behaviors. Project Page: https://MA-Bench.github.io

Kun Li, Jihao Gu, Fei Wang +3

CVLab, College of Information Technology, United Arab Emirates University · University College London · Hefei University of Technology +2

MLLM EvaluationMultimodal Large Language ModelsMultimodal Model EvaluationHuman Activity Recognition
AskPDF
Mar 27, 2026

Rocks, Pebbles and Sand: Modality-aware Scheduling for Multimodal Large Language Model Inference

Multimodal Large Language Models (MLLMs) power platforms like ChatGPT, Gemini, and Copilot, enabling rich interactions across text, images, and videos. Unlike text-only workloads, multimodal requests are inherently heterogeneous, differing by orders of magnitude in latency and memory footprint. Existing LLM serving systems, designed for homogeneous text workloads, struggle under this heterogeneity: large image and video requests monopolize GPU resources, causing severe head-of-line blocking and degrading the responsiveness of lightweight, latency-critical text queries. Recent multimodal serving systems mitigate this interference through disaggregation, isolating encoding and inference stages across GPUs. While effective at large scale, such designs introduce severe GPU underutilization in more practical, moderate-scale deployments. In response, we build RPS-Serve a modality-aware scheduling framework that leverages the extreme heterogeneity in multimodal workloads and introduces a simple abstraction: videos behave like rocks, images like pebbles, and text like sand. Then, it lets sand flow quickly through pebbles and rocks, maximizing GPU utilization while ensuring fast responsiveness for lightweight requests and controlled execution for heavier ones. To achieve this, RPS-Serve estimates the GPU admission cost of each request, classifies requests into rocks, pebbles, and sand using lightweight clustering, and applies priority-based scheduling with aging to balance performance across modalities. Evaluation across state-of-the-art MLLMs shows that RPS-Serve increases GPU utilization by 65% and reduces TTFT by up to 9.3x, on average across models, compared to disaggregation-based multi-modal systems.

Konstantinos Papaioannou, Thaleia Dimitra Doudali

IMDEA Software Institute, Universidad Politécnica de Madrid

LLM ServingLLM Inference SchedulingEfficient Multimodal InferenceMultimodal Large Language Models
AskPDF
Mar 27, 2026

Reflect to Inform: Boosting Multimodal Reasoning via Information-Gain-Driven Verification

Multimodal Large Language Models (MLLMs) achieve strong multimodal reasoning performance, yet we identify a recurring failure mode in long-form generation: as outputs grow longer, models progressively drift away from image evidence and fall back on textual priors, resulting in ungrounded reasoning and hallucinations. Interestingly, Based on attention analysis, we find that MLLMs have a latent capability for late-stage visual verification that is present but not consistently activated. Motivated by this observation, we propose Visual Re-Examination (VRE), a self-evolving training framework that enables MLLMs to autonomously perform visual introspection during reasoning without additional visual inputs. Rather than distilling visual capabilities from a stronger teacher, VRE promotes iterative self-improvement by leveraging the model itself to generate reflection traces, making visual information actionable through information gain. Extensive experiments across diverse multimodal benchmarks demonstrate that VRE consistently improves reasoning accuracy and perceptual reliability, while substantially reducing hallucinations, especially in long-chain settings. Code is available at https://github.com/Xiaobu-USTC/VRE.

Shuai Lv, Chang Liu, Feng Tang +5

University of Science and Technology of China, Hefei, China · Huawei Foundation Model Department, Shanghai, China · The Chinese University of Hong Kong, Hong Kong, China +1

Multimodal Large Language ModelsMultimodal ReasoningVLM Hallucination MitigationVisual Grounding
AskPDF
Mar 27, 2026

FairLLaVA: Fairness-Aware Parameter-Efficient Fine-Tuning for Large Vision-Language Assistants

While powerful in image-conditioned generation, multimodal large language models (MLLMs) can display uneven performance across demographic groups, highlighting fairness risks. In safety-critical clinical settings, such disparities risk producing unequal diagnostic narratives and eroding trust in AI-assisted decision-making. While fairness has been studied extensively in vision-only and language-only models, its impact on MLLMs remains largely underexplored. To address these biases, we introduce FairLLaVA, a parameter-efficient fine-tuning method that mitigates group disparities in visual instruction tuning without compromising overall performance. By minimizing the mutual information between target attributes, FairLLaVA regularizes the model's representations to be demographic-invariant. The method can be incorporated as a lightweight plug-in, maintaining efficiency with low-rank adapter fine-tuning, and provides an architecture-agnostic approach to fair visual instruction following. Extensive experiments on large-scale chest radiology report generation and dermoscopy visual question answering benchmarks show that FairLLaVA consistently reduces inter-group disparities while improving both equity-scaled clinical performance and natural language generation quality across diverse medical imaging modalities. Code can be accessed at https://github.com/bhosalems/FairLLaVA.

Mahesh Bhosale, Abdul Wasi, Shantam Srivastava +5

University at Buffalo · University of Kashmir · Accenture +1

Multimodal Large Language ModelsMedical VLMs
AskPDF
Mar 26, 2026

GazeQwen: Lightweight Gaze-Conditioned LLM Modulation for Streaming Video Understanding

Current multimodal large language models (MLLMs) cannot effectively utilize eye-gaze information for video understanding, even when gaze cues are supplied via visual overlays or text descriptions. We introduce GazeQwen, a parameter efficient approach that equips an open-source MLLM with gaze awareness through hidden-state modulation. At its core is a compact gaze resampler (~1-5 M trainable parameters) that encodes V-JEPA 2.1 video features together with fixation-derived positional encodings and produces additive residuals injected into selected LLM decoder layers via forward hooks. An optional second training stage adds low-rank adapters (LoRA) to the LLM for tighter integration. Evaluated on all 10 tasks of the StreamGaze benchmark, GazeQwen reaches 63.9% accuracy, a +16.1 point gain over the same Qwen2.5-VL-7B backbone with gaze as visual prompts and +10.5 points over GPT-4o, the highest score among all open-source and proprietary models tested. These results suggest that learning where to inject gaze within an LLM is more effective than scaling model size or engineering better prompts. All code and checkpoints are available at https://github.com/phamtrongthang123/gazeqwen .

Trong Thang Pham, Hien Nguyen, Ngan Le

University of Arkansas · University of Houston

Multimodal Large Language ModelsStreaming Video UnderstandingParameter-Efficient Fine-Tuning
AskPDF
Mar 26, 2026

Demographic Fairness in Multimodal LLMs: A Benchmark of Gender and Ethnicity Bias in Face Verification

Multimodal Large Language Models (MLLMs) have recently been explored as face verification systems that determine whether two face images are of the same person. Unlike dedicated face recognition systems, MLLMs approach this task through visual prompting and rely on general visual and reasoning abilities. However, the demographic fairness of these models remains largely unexplored. In this paper, we present a benchmarking study that evaluates nine open-source MLLMs from six model families, ranging from 2B to 8B parameters, on the IJB-C and RFW face verification protocols across four ethnicity groups and two gender groups. We measure verification accuracy with the Equal Error Rate and True Match Rate at multiple operating points per demographic group, and we quantify demographic disparity with four FMR-based fairness metrics. Our results show that FaceLLM-8B, the only face-specialised model in our study, substantially outperforms general-purpose MLLMs on both benchmarks. The bias patterns we observe differ from those commonly reported for traditional face recognition, with different groups being most affected depending on the benchmark and the model. We also note that the most accurate models are not necessarily the fairest and that models with poor overall accuracy can appear fair simply because they produce uniformly high error rates across all demographic groups.

Ünsal Öztürk, Hatef Otroshi Shahreza, Sébastien Marcel

Idiap Research Institute, Switzerland

Algorithmic FairnessMultimodal Large Language ModelsFace RecognitionMultimodal Model Evaluation
AskPDF
Mar 26, 2026

AnyDoc: Enhancing Document Generation via Large-Scale HTML/CSS Data Synthesis and Height-Aware Reinforcement Optimization

Document generation has gained growing attention in the field of AI-driven content creation. In this work, we push its boundaries by introducing AnyDoc, a framework capable of handling multiple generation tasks across a wide spectrum of document categories, all represented in a unified HTML/CSS format. To overcome the limited coverage and scale of existing human-crafted document datasets, AnyDoc first establishes a scalable data synthesis pipeline to automatically generate documents in HTML/CSS form. This pipeline yields DocHTML, a large-scale dataset containing 265,206 document samples, while spanning 111 categories and 32 distinct styles. Additionally, all documents are equipped with comprehensive metadata, including design intentions, HTML/CSS source code, visual assets, and rendered screenshots. Building on the curated dataset, AnyDoc fine-tunes multi-modal large language models (MLLMs) to achieve three practical document generation tasks: intention-to-document, document derendering, and element-to-document. To address the content overflow issue observed during fine-tuning, AnyDoc further incorporates a height-aware reinforcement learning (HARL) post-training procedure. By defining a reward function based on the difference between predicted and target document heights, overflow is penalized and gradually mitigated during HARL, thereby enhancing overall performance. Qualitative and quantitative experiments demonstrate that AnyDoc outperforms both general-purpose MLLMs and task-specific baselines across all three tasks.

Jiawei Lin, Wanrong Zhu, Vlad I Morariu +1

Xi’an Jiaotong University · Adobe Research

Text GenerationSynthetic Data Generation for Language ModelsLLM Fine-TuningMultimodal Large Language ModelsRL for Generative Models
AskPDF
Mar 26, 2026

Bridging Perception and Reasoning: Token Reweighting for RLVR in Multimodal LLMs

Extending Reinforcement Learning with Verifiable Rewards (RLVR) to multimodal large language models (MLLMs) faces a fundamental challenge: their responses inherently interleave perception-related tokens, which ground visual content, with reasoning-related tokens, which construct reasoning chains. These token types instantiate distinct yet interdependent capacities -- visual grounding and symbolic reasoning -- making isolated optimization insufficient. Through token-level empirical analysis, we demonstrate that optimizing either perception- or reasoning-only tokens consistently underperforms full optimization, underscoring their inherent coupling. To address this, we propose a plug-and-play Token-Reweighting (ToR) strategy that explicitly models this interdependence by identifying critical tokens of both types and dynamically reweighting them during RLVR training. Applied on top of existing methods (e.g., GRPO and DAPO), ToR delivers consistent performance gains across multiple multi-modal reasoning benchmarks, achieving state-of-the-art performance with both accurate visual grounding and coherent reasoning.

Jinda Lu, Junkang Wu, Jinghan Li +6

University of Science and Technology of China · Peking University · Independent Researcher

RL for VLMsVLM ReasoningMultimodal Large Language ModelsReinforcement Learning with Verifiable Rewards
AskPDF
Mar 25, 2026

ReLope: From Hidden-State Probing to a Decision Module for Multimodal LLM Routing

Routing balances performance and cost in hybrid systems by escalating selected queries from a lightweight model to a powerful but expensive one. Hidden-state probes provide an effective routing signal by predicting the small model's correctness from hidden states it already computes. Although effective in text-only settings, their behavior on multimodal LLMs (MLLMs) is less understood. We find that correctness is harder to predict from hidden states when questions are paired with images rather than with captions of those images. Two training-free dependence measures, HSIC and label CKA, show the same pattern. We first investigate whether changing hidden states used by the probe improves correctness prediction for MLLMs. We propose Attention Probe, which learns an attention query to pool frozen token states into a routing feature and improves over the standard probe. However, pooling only recombines hidden states computed for answer generation, not for deciding whether the answer should be trusted. We therefore build a decision layer inside the existing MLLM and train it for the routing decision. The resulting router, ReLope (KL-Regularized LoRA Probe), adapts a separate copy of one MLLM layer with LoRA and regularizes it with a KL-penalized stochastic bottleneck. This decision branch shares the frozen lower layers and leaves the model's answer-generation path unchanged. ReLope attains the highest correctness prediction AUC on all five benchmarks and three backbones we evaluate and improves the accuracy versus cost trade-off at negligible overhead. These results suggest a new perspective and a practical technical path on routing: rather than reading decisions off states built for generation, routing decisions can be learned by a dedicated decision layer inside the model. Code: https://github.com/Spinozaaa/ReLope

Yaopei Zeng, Congchao Wang, Blake JianHang Chen +1

Pennsylvania State University · Google · Amazon AGI

Multimodal Large Language ModelsLLM RoutingLanguage Model Probing
AskPDF
Mar 25, 2026

ReDiPrune: Relevance-Diversity Pre-Projection Token Pruning for Efficient Multimodal LLMs

Recent multimodal large language models are computationally expensive because Transformers must process a large number of visual tokens. We present ReDiPrune, a training-free token pruning method applied before the vision-language projector, where visual features remain rich and discriminative. Unlike post-projection pruning methods that operate on compressed representations, ReDiPrune selects informative tokens directly from vision encoder outputs, preserving fine-grained spatial and semantic cues. Each token is scored by a lightweight rule that jointly consider text-conditioned relevance and max-min diversity, ensuring the selected tokens are both query-relevant and non-redundant. ReDiPrune is fully plug-and-play, requiring no retraining or architectural modifications, and can be seamlessly inserted between the encoder and projector. Across four video and five image benchmarks, it consistently improves the accuracy-efficiency trade-off. For example, on EgoSchema with LLaVA-NeXT-Video-7B, retaining only 15% of visual tokens yields a +2.0% absolute accuracy gain while reducing computation by more than 6×6\times in TFLOPs. Code is available at https://github.com/UA-CVML/ReDiPrune.

An Yu, Ting Yu Tsai, Zhenfei Zhang +3

University at Albany, SUNY · Peking University

Efficient VLM InferenceToken PruningVisual Token PruningLVLM PruningMultimodal Large Language ModelsMultimodal Token Compression
AskPDF
Mar 25, 2026

Lightweight Multimodal LLM-Enabled Cost-Effective Defect Grading of Power Transmission Equipment

Defect grading of power transmission equipment (DGPTE) is crucial to the stability of electric energy transmission. Although existing machine learning methods exhibit strong capabilities in defect detection, they are plagued by difficulties in integrating expert experience and facing class imbalance in more refined defect grading field. To address this issue, this paper introduces a novel defect grading framework based on multimodal large language model (MLLM). Specifically, this approach maximizes the commercial MLLMs' potential of DGPTE through in-context learning and obtains the state-of-te-art (SOTA) model. By sending a secondary request to this model, a small number of chain of thought-based question-answer pairs (Q&As) are generated, which effectively reduces the cost of manual annotation. In this way, these high-quality interpretable Q&As are used to train Qwen3-VL-8B via Low-Rank Adaption-based supervised fine-tuning (SFT). Experimental results on three DGPTE tasks demonstrate that fine-tuning only the language model layer yields the SOTA performance. Furthermore, multi-task joint fine-tuning verifies the feasibility of handling multiple grading tasks within only a single lightweight MLLM.

Tao Wang, Lipeng Zhu, Jiayong Li +2

Multimodal Large Language ModelsLLM Fine-Tuning
AskPDF
Mar 25, 2026

Video-Only ToM: Enhancing Theory of Mind in Multimodal Large Language Models

As large language models (LLMs) continue to advance, there is increasing interest in their ability to infer human mental states and demonstrate a human-like Theory of Mind (ToM). Most existing ToM evaluations, however, are centered on text-based inputs, while scenarios relying solely on visual information receive far less attention. This leaves a gap, since real-world human-AI interaction typically requires multimodal understanding. In addition, many current methods regard the model as a black box and rarely probe how its internal attention behaves in multiple-choice question answering (QA). The impact of LLM hallucinations on such tasks is also underexplored from an interpretability perspective. To address these issues, we introduce VisionToM, a vision-oriented intervention framework designed to strengthen task-aware reasoning. The core idea is to compute intervention vectors that align visual representations with the correct semantic targets, thereby steering the model's attention through different layers of visual features. This guidance reduces the model's reliance on spurious linguistic priors, leading to more reliable multimodal language model (MLLM) outputs and better QA performance. Experiments on the EgoToM benchmark-an egocentric, real-world video dataset for ToM with three multiple-choice QA settings-demonstrate that our method substantially improves the ToM abilities of MLLMs. Furthermore, results on an additional open-ended generation task show that VisionToM enables MLLMs to produce free-form explanations that more accurately capture agents' mental states, pushing machine-human collaboration toward greater alignment.

Siqi Liu, Xinyang Li, Bochao Zou +3

University of Science and Technology Beijing

Theory of MindMultimodal Large Language ModelsVLM ReasoningVLM Interpretability
AskPDF
Mar 25, 2026

When Understanding Becomes a Risk: Authenticity and Safety Risks in the Emerging Image Generation Paradigm

Recently, multimodal large language models (MLLMs) have emerged as a unified paradigm for language and image generation. Compared with diffusion models, MLLMs possess a much stronger capability for semantic understanding, enabling them to process more complex textual inputs and comprehend richer contextual meanings. However, this enhanced semantic ability may also introduce new and potentially greater safety risks. Taking diffusion models as a reference point, we systematically analyze and compare the safety risks of emerging MLLMs along two dimensions: unsafe content generation and fake image synthesis. Across multiple unsafe generation benchmark datasets, we observe that MLLMs tend to generate more unsafe images than diffusion models. This difference partly arises because diffusion models often fail to interpret abstract prompts, producing corrupted outputs, whereas MLLMs can comprehend these prompts and generate unsafe content. For current advanced fake image detectors, MLLM-generated images are also notably harder to identify. Even when detectors are retrained with MLLMs-specific data, they can still be bypassed by simply providing MLLMs with longer and more descriptive inputs. Our measurements indicate that the emerging safety risks of the cutting-edge generative paradigm, MLLMs, have not been sufficiently recognized, posing new challenges to real-world safety.

Ye Leng, Junjie Chu, Mingjie Li +5

CISPA Helmholtz Center for Information Security · Xi’an Jiaotong University · Flexera

MLLM SafetyImage GenerationLLM SafetyMultimodal Large Language ModelsT2I Generation EvaluationAI-Generated Image Detection
AskPDF
Mar 25, 2026

A Step Toward Federated Pretraining of Multimodal Large Language Models

The rapid evolution of Multimodal Large Language Models (MLLMs) is bottlenecked by the saturation of high-quality public data, while vast amounts of diverse multimodal data remain inaccessible in privacy-sensitive silos. Federated Learning (FL) offers a promising solution to unlock these distributed resources, but existing research focuses predominantly on fine-tuning, leaving the foundational pre-training phase largely unexplored. In this paper, we formally introduce the Federated MLLM Alignment (Fed-MA) task, a lightweight pre-training paradigm that freezes the vision encoder and LLM while collaboratively training the cross-modal projector. We identify two critical challenges in this setting: (i) parameter interference in aggregating local projectors; and (ii) gradient oscillations in one-pass collaborative SGD. To address these challenges, we propose Fed-CMP, a pioneering framework for federated MLLM pre-training. Fed-CMP employs Canonical Reliability-Aware Aggregation, which constructs a canonical space to decompose client projectors into a shared alignment basis and client-specific coefficients, then performs reliability-weighted fusion to suppress parameter interference. Furthermore, Fed-CMP introduces Orthogonality-Preserved Momentum, which applies momentum to the shared alignment basis via orthogonal projection, accumulating historical optimization directions while preserving geometric structure. We construct four federated pre-training scenarios based on public datasets, and extensive experiments validate that Fed-CMP significantly outperforms existing baselines.

Baochen Xiong, Yifan Xu, Xiaoshan Yang +3

MAIS, Institute of Automation, Chinese Academy of Sciences · Pengcheng Laboratory · School of Artificial Intelligence, University of Chinese Academy of Sciences (UCAS) +1

Multimodal Federated LearningMultimodal PretrainingMultimodal Large Language ModelsFederated Learning Aggregation
AskPDF
Mar 25, 2026

A3^3: Towards Advertising Aesthetic Assessment

Advertising images significantly impact commercial conversion rates and brand equity, yet current evaluation methods rely on subjective judgments, lacking scalability, standardized criteria, and interpretability. To address these challenges, we present A^3 (Advertising Aesthetic Assessment), a comprehensive framework encompassing four components: a paradigm (A^3-Law), a dataset (A^3-Dataset), a multimodal large language model (A^3-Align), and a benchmark (A^3-Bench). Central to A^3 is a theory-driven paradigm, A^3-Law, comprising three hierarchical stages: (1) Perceptual Attention, evaluating perceptual image signals for their ability to attract attention; (2) Formal Interest, assessing formal composition of image color and spatial layout in evoking interest; and (3) Desire Impact, measuring desire evocation from images and their persuasive impact. Building on A^3-Law, we construct A^3-Dataset with 120K instruction-response pairs from 30K advertising images, each richly annotated with multi-dimensional labels and Chain-of-Thought (CoT) rationales. We further develop A^3-Align, trained under A^3-Law with CoT-guided learning on A^3-Dataset. Extensive experiments on A^3-Bench demonstrate that A^3-Align achieves superior alignment with A^3-Law compared to existing models, and this alignment generalizes well to quality advertisement selection and prescriptive advertisement critique, indicating its potential for broader deployment. Dataset, code, and models can be found at: https://github.com/euleryuan/A3-Align.

Kaiyuan Ji, Yixuan Gao, Lu Sun +7

Shanghai Artificial Intelligence Laboratory · School of Information and Electronic Engineering, East China Normal University · Institute of Image Communication and Network Engineering, Shanghai Jiao Tong University +1

Image Quality AssessmentMultimodal Large Language Models
AskPDF
Mar 25, 2026

Thinking with Tables: Enhancing Multi-Modal Tabular Understanding via Neuro-Symbolic Reasoning

Multimodal Large Language Models (MLLMs) have demonstrated remarkable reasoning capabilities across modalities such as images and text. However, tabular data, despite being a critical real-world modality, remains relatively underexplored in multimodal learning. In this paper, we focus on the task of Tabular-Vision Multi-Modal Understanding (TVMU) and identify three core challenges: (1) high structural variability and data incompleteness in tables, (2) implicit and complex feature dependencies, and (3) significant heterogeneity in problem-solving pipelines across downstream tasks. To address these issues, we propose Thinking with Tables (TWT). TWT employs a program-aided code-based neuro-symbolic reasoning mechanism that facilitates key operations, such as information extraction and element modeling, by interacting with external environments. We evaluate TWT on eight representative datasets. Experimental results demonstrate that TWT consistently outperforms existing baselines by an average of 10% in accuracy, achieving performance comparable to, or even surpassing, proprietary commercial SOTA LLMs on TVMU tasks. Models and codes are available at https://github.com/kunyang-YU/Thinking-with-Tables

Kun-Yang Yu, Zhi Zhou, Shi-Yu Tian +6

State Key Laboratory of Novel Software Technology, Nanjing University, China · School of Artificial Intelligence, Nanjing University, China · School of Intelligence Science and Technology, Nanjing University, China

Multimodal UnderstandingMultimodal Large Language ModelsMultimodal ReasoningLLM ReasoningNeuro-Symbolic Reasoning
AskPDF
Mar 25, 2026

Towards Real-World Document Parsing via Realistic Scene Synthesis and Document-Aware Training

Document parsing has recently advanced with multimodal large language models (MLLMs) that directly map document images to structured outputs. Traditional cascaded pipelines depend on precise layout analysis and often fail under casually captured or non-standard conditions. Although end-to-end approaches mitigate this dependency, they still exhibit repetitive, hallucinated, and structurally inconsistent predictions - primarily due to the scarcity of large-scale, high-quality full-page (document-level) end-to-end parsing data and the lack of structure-aware training strategies. To address these challenges, we propose a data-training co-design framework for robust end-to-end document parsing. A Realistic Scene Synthesis strategy constructs large-scale, structurally diverse full-page end-to-end supervision by composing layout templates with rich document elements, while a Document-Aware Training Recipe introduces progressive learning and structure-token optimization to enhance structural fidelity and decoding stability. We further build Wild-OmniDocBench, a benchmark derived from real-world captured documents for robustness evaluation. Integrated into a 1B-parameter MLLM, our method achieves superior accuracy and robustness across both scanned/digital and real-world captured scenarios. All models, data synthesis pipelines, and benchmarks will be publicly released to advance future research in document understanding.

Gengluo Li, Pengyuan Lyu, Chengquan Zhang +7

Institute of Information Engineering, Chinese Academy of Sciences · University of Chinese Academy of Sciences · Tencent +2

Document ParsingMultimodal Large Language ModelsSynthetic Data Generation
AskPDF
Mar 24, 2026

3DCity-LLM: Empowering Multi-modality Large Language Models for 3D City-scale Perception and Understanding

While multi-modality large language models excel in object-centric or indoor scenarios, scaling them to 3D city-scale environments remains a formidable challenge. To bridge this gap, we propose 3DCity-LLM, a unified framework designed for 3D city-scale vision-language perception and understanding. 3DCity-LLM employs a coarse-to-fine feature encoding strategy comprising three parallel branches for target object, inter-object relationship, and global scene. To facilitate large-scale training, we introduce 3DCity-LLM-1.2M dataset that comprises approximately 1.2 million high-quality samples across seven representative task categories, ranging from fine-grained object analysis to multi-faceted scene planning. This strictly quality-controlled dataset integrates explicit 3D numerical information and diverse user-oriented simulations, enriching the question-answering diversity and realism of urban scenarios. Furthermore, we apply a multi-dimensional protocol based on text-similarity metrics and LLM-based semantic assessment to ensure faithful and comprehensive evaluations for all methods. Extensive experiments on two benchmarks demonstrate that 3DCity-LLM significantly outperforms existing state-of-the-art methods, offering a promising and meaningful direction for advancing spatial reasoning and urban intelligence. The source code and dataset are available at https://github.com/SYSU-3DSTAILab/3D-City-LLM.

Yiping Chen, Jinpeng Li, Wenyu Ke +6

School of Geospatial Engineering and Science, Sun Yat-sen University, Zhuhai, 519082, China. · College of Electronic Science, National University of Defense Technology, Changsha, 410000, China. · Department of Civil and Environmental Engineering, Norwegian University of Science and Technology, Trondheim, Norway. +1

3D VLMs3D Scene UnderstandingMultimodal Large Language Models3D VQA
AskPDF
Mar 24, 2026

Unleashing Spatial Reasoning in Multimodal Large Language Models via Textual Representation Guided Reasoning

Existing Multimodal Large Language Models (MLLMs) struggle with 3D spatial reasoning, as they fail to construct structured abstractions of the 3D environment depicted in video inputs. To bridge this gap, drawing inspiration from cognitive theories of allocentric spatial reasoning, we investigate how to enable MLLMs to model and reason over text-based spatial representations of video. Specifically, we introduce Textual Representation of Allocentric Context from Egocentric Video (TRACE), a prompting method that induces MLLMs to generate text-based representations of 3D environments as intermediate reasoning traces for more accurate spatial question answering. TRACE encodes meta-context, camera trajectories, and detailed object entities to support structured spatial reasoning over egocentric videos. Extensive experiments on VSI-Bench and OST-Bench demonstrate that TRACE yields notable and consistent improvements over prior prompting strategies across a diverse range of MLLM backbones, spanning different parameter scales and training schemas. We further present ablation studies to validate our design choices, along with detailed analyses that probe the bottlenecks of 3D spatial reasoning in MLLMs.

Jiacheng Hua, Yishu Yin, Yuhang Wu +3

College of AI, Tsinghua University, Beijing, China · Shanghai Artificial Intelligence Laboratory, Shanghai, China · The University of Tokyo, Tokyo, Japan

3D Spatial ReasoningMultimodal Large Language ModelsEgocentric Video QA
AskPDF
Mar 24, 2026

SIMART: Decomposing Monolithic Meshes into Sim-ready Articulated Assets via MLLM

High-quality articulated 3D assets are indispensable for embodied AI and physical simulation, yet 3D generation still focuses on static meshes, leaving a gap in "sim-ready" interactive objects. Most recent articulated object creation methods rely on multi-stage pipelines that accumulate errors across decoupled modules. Alternatively, unified MLLMs offer a single-stage path to joint static asset understanding and sim-ready asset generation. However dense voxel-based 3D tokenization yields long 3D token sequences and high memory overhead, limiting scalability to complex articulated objects. To address this, we propose SIMART, a unified MLLM framework that jointly performs part-level decomposition and kinematic prediction. By introducing a Sparse 3D VQ-VAE, SIMART reduces token counts by 70% vs. dense voxel tokens, enabling high-fidelity multi-part assemblies. SIMART achieves state-of-the-art performance on PartNet-Mobility and in-the-wild AIGC datasets, and enables physics-based robotic simulation.

Chuanrui Zhang, Minghan Qin, Yuang Wang +3

ByteDance Seed · NTU

Articulated 3D Asset Generation3D Asset GenerationPart-Level 3D GenerationMultimodal Large Language Models3D Part SegmentationRobotics Simulation
AskPDF