Multimodal Reasoning

Follow this topic Ask about this topic

Multimodal reasoning studies how models perceive information across modalities, connect that evidence to knowledge, and use it to answer questions, plan actions, or explain decisions. Common approaches include instruction tuning and chain-of-thought training, reinforcement learning with verifiable rewards, tool-using agents, and structured methods such as graphs or symbolic solvers. Benchmarks such as MathVista and MMMU test visual problem solving, while video, document, scientific, and medical tasks probe reasoning across time and specialized evidence. A central challenge is ensuring that correct answers actually depend on relevant multimodal evidence rather than language shortcuts; grounding, cross-modal integration, and reliable generalization remain difficult.

Latest papers

512 papers47 papers in the last four weeks, down 10% on the four weeks before. 0.3% of all new papers.

Apr 16, 2026

Knowing When Not to Answer: Evaluating Abstention in Multimodal Reasoning Systems

Effective abstention (EA), recognizing evidence insufficiency and refraining from answering, is critical for reliable multimodal systems. Yet existing evaluation paradigms for vision-language models (VLMs) and multi-agent systems (MAS) assume answerability, pushing models to always respond. Abstention has been studied in text-only settings but remains underexplored multimodally; current benchmarks either ignore unanswerability or rely on coarse methods that miss realistic failure modes. We introduce MM-AQA, a benchmark that constructs unanswerable instances from answerable ones via transformations along two axes: visual modality dependency and evidence sufficiency. Evaluating three frontier VLMs spanning closed and open-source models and two MAS architectures across 2079 samples, we find: (1) under standard prompting, VLMs rarely abstain; even simple confidence baselines outperform this setup, (2) MAS improves abstention but introduces an accuracy-abstention trade-off, (3) sequential designs match or exceed iterative variants, suggesting the bottleneck is miscalibration rather than reasoning depth, and (4) models abstain when image or text evidence is absent, but attempt reconciliation with degraded or contradictory evidence. Effective multimodal abstention requires abstention-aware training rather than better prompting or more agents.

Nishanth Madhusudhan, Vikas Yadav, Alexandre Lacoste

ServiceNow Research

Multimodal QA Multimodal Reasoning Multimodal Agents Multimodal Model Evaluation Multimodal Understanding

Apr 16, 2026

Chain of Modality: From Static Fusion to Dynamic Orchestration in Omni-MLLMs

Omni-modal Large Language Models (Omni-MLLMs) are designed to reason over diverse sensory streams within a unified model. However, we find that multimodal reasoning is not solely determined by sensory evidence, but also by how modality information is organized. Through systematic analysis across diverse reasoning scenarios, we show that different modality organizations exhibit distinct advantages and limitations, and no single topology is universally optimal. Rather than proposing a general-purpose improvement to Omni-MLLM accuracy, we ask a narrower question: can these topology-induced failures be systematically identified and corrected? Motivated by this, we propose Chain of Modality (CoM), a framework that dynamically reorganizes multimodal topology during inference and learns adaptive organization strategies. Experiments across five benchmarks, diverse architectures, and model scales show that CoM reliably recovers topology-sensitive failures, through both training-free planning and lightweight Planner-SFT, while preserving performance on the full benchmark.

Ziyang Luo, Nian Liu, Junwei Han

Northwestern Polytechnical University Xi’an, Shaanxi, China

Multimodal Large Language Models Multimodal Reasoning Multimodal Fusion Multimodal Understanding

Apr 15, 2026

Seek-and-Solve: Benchmarking MLLMs for Visual Clue-Driven Reasoning in Daily Scenarios

Daily scenarios are characterized by visual richness, requiring Multimodal Large Language Models (MLLMs) to filter noise and identify decisive visual clues for accurate reasoning. Yet, current benchmarks predominantly aim at evaluating MLLMs' pre-existing knowledge or perceptual understanding, often neglecting the critical capability of reasoning. To bridge this gap, we introduce DailyClue, a benchmark designed for visual clue-driven reasoning in daily scenarios. Our construction is guided by two core principles: (1) strict grounding in authentic daily activities, and (2) challenging query design that necessitates more than surface-level perception. Instead of simple recognition, our questions compel MLLMs to actively explore suitable visual clues and leverage them for subsequent reasoning. To this end, we curate a comprehensive dataset spanning four major daily domains and 16 distinct subtasks. Comprehensive evaluation across MLLMs and agentic models underscores the formidable challenge posed by our benchmark. Our analysis reveals several critical insights, emphasizing that the accurate identification of visual clues is essential for robust reasoning.

Xiaomin Li, Tala Wang, Zichen Zhong +7

Dalian University of Technology · WeChat, Tencent Inc. · Tsinghua University

VLM Evaluation Multimodal Reasoning Visually Grounded Reasoning

Apr 15, 2026

MERRIN: A Benchmark for Multimodal Evidence Retrieval and Reasoning in Noisy Web Environments

Motivated by the underspecified, multi-hop nature of search queries and the multimodal, heterogeneous, and often conflicting nature of real-world web results, we introduce MERRIN (Multimodal Evidence Retrieval and Reasoning in Noisy Web Environments), a human-annotated benchmark for evaluating search-augmented agents. MERRIN measures AI agents' ability to identify relevant modalities, retrieve multimodal evidence, and perform multi-hop reasoning over noisy web sources. It differs from prior work in three important aspects: (1) using natural language queries without explicit modality cues, (2) incorporating underexplored modalities such as video and audio, and (3) requiring the retrieval of complex, often noisy or conflicting multimodal evidence during web search. We evaluate diverse search agents powered by ten models, including strong closed-source models (e.g., GPT-5.4-mini, Gemini 3/3.1 Flash/Pro) and open-weight models (Qwen3-4B/30B/235B), across three search settings (no search, native search, and agentic search). Our results show that MERRIN is highly challenging: the average accuracy across all agents is 22.3%, with the best-performing agent reaching only 40.1%. We further observe that while stronger agents like Gemini Deep Research achieve higher performance, gains are modest due to over-exploration; they take more steps and use more tools, but are often distracted by conflicting or partially relevant web content, leading to incorrect answers. Compared to humans, these agents consume more resources yet achieve lower accuracy, largely due to inefficient source selection and an overreliance on text modalities. These findings highlight the need for search agents capable of robust search and reasoning across diverse modalities in noisy web environments, making MERRIN a valuable testbed for evaluating such capabilities.

Han Wang, David Wan, Hyunji Lee +6

UNC Chapel Hill · Virginia Tech · University of Texas at Austin

AI Agent Benchmarks Multimodal Search Agents Multimodal IR Multimodal Reasoning Multi-Hop Reasoning

Apr 14, 2026

Medical thinking with multiple images

Large language models perform well on many medical QA benchmarks, but real clinical reasoning often requires integrating evidence across multiple images rather than interpreting a single view. We introduce MedThinkVQA, an expert-annotated benchmark for thinking with multiple images, where models must interpret each image, combine cross-view evidence, and answer diagnostic questions with intermediate supervision and step-level evaluation. The dataset contains 8,067 cases, including 720 test cases, with an average of 6.62 images per case, substantially denser than prior work, whose expert-level benchmarks use at most 1.43 images per case. On the test set, the best closed-source models, Claude-4.6-Opus, Gemini-3-Pro, and GPT-5.2-xhigh, reach only 57.2%, 55.3%, and 54.9% accuracy, while GPT-5-mini and GPT-5-nano reach 39.7% and 30.8%. Strong open-source models lag behind, led by Qwen3.5-397B-A17B at 52.2% and Qwen3.5-27B at 50.6%. Further analysis identifies grounded multi-image reasoning as the main bottleneck: models often fail to extract, align, and compose evidence across views before higher-level inference can help. Providing expert single-image cues and cross-image summaries improves performance, whereas replacing them with self-generated intermediates reduces accuracy. Step-level analysis shows that over 70% of errors arise from image reading and cross-view integration. Scaling results further show that additional inference-time computation helps only when visual grounding is already reliable; when early evidence extraction is weak, longer reasoning yields limited or unstable gains and can amplify misread cues. These results suggest that the key challenge is not reasoning length alone, but reliable mechanisms for grounding, aligning, and composing distributed evidence across real-world multimodal clinical inputs.

Zonghai Yao, Benlu Wang, Yifan Zhang +8

Manning College of Information and Computer Sciences, UMass Amherst · Center for Healthcare Organization and Implementation Research, VA Bedford Health Care · Department of Computer Science, Yale University +2

Medical VQA Multimodal Reasoning Multi-Image Reasoning Medical Image Benchmarks Medical Image Grounding

Apr 14, 2026

Towards Long-horizon Agentic Multimodal Search

Multimodal deep search agents have shown great potential in solving complex tasks by iteratively collecting textual and visual evidence. However, managing the heterogeneous information and high token costs associated with multimodal inputs over long horizons remains a critical challenge, as existing methods often suffer from context explosion or the loss of crucial visual signals. To address this, we propose a novel Long-horizon MultiModal deep search framework, named LMM-Searcher, centered on a file-based visual representation mechanism. By offloading visual assets to an external file system and mapping them to lightweight textual identifiers (UIDs), our approach mitigates context overhead while preserving multimodal information for future access. We equip the agent with a tailored fetch-image tool, enabling a progressive, on-demand visual loading strategy for active perception. Furthermore, we introduce a data synthesis pipeline designed to generate queries requiring complex cross-modal multi-hop reasoning. Using this pipeline, we distill 12K high-quality trajectories to fine-tune Qwen3-VL-Thinking-30A3B into a specialized multimodal deep search agent. Extensive experiments across four benchmarks demonstrate that our method successfully scales to 100-turn search horizons, achieving state-of-the-art performance among open-source models on challenging long-horizon benchmarks like MM-BrowseComp and MMSearch-Plus, while also exhibiting strong generalizability across different base models. Our code will be released in https://github.com/RUCAIBox/LMM-Searcher.

Yifan Du, Zikang Liu, Jinbiao Peng +5

Gaoling School of Artificial Intelligence, Renmin University of China. · City University of Hong Kong.

Multimodal Search Agents Long-Horizon LLM Agents Multimodal Reasoning

Apr 14, 2026

MISID: A Multimodal Multi-turn Dataset for Complex Intent Recognition in Strategic Deception Games

Understanding human intent in complex multi-turn interactions remains a fundamental challenge in human-computer interaction and behavioral analysis. While existing intent recognition datasets focus mainly on single utterances or simple dialogues, real-world scenarios often involve sophisticated strategic interactions where participants must maintain complex deceptive narratives over extended periods. To address this gap, we introduce MISID, a comprehensive multimodal, multi-turn, and multi-participant benchmark for intent recognition. Sourced from high-stakes social strategy games, MISID features a fine-grained, two-tier multi-dimensional annotation scheme tailored for long-context discourse analysis and evidence-based causal tracking. Our systematic evaluation of state-of-the-art Multimodal Large Language Models (MLLMs) on MISID reveals critical deficiencies in complex scenarios, including text-prior visual hallucination, impaired cross-modal synergy, and limited capacity in chaining causal cues. Consequently, we propose FRACTAM as a baseline framework. Using a ``Decouple-Anchor-Reason'' paradigm, FRACTAM reduces text bias by extracting pure unimodal factual representations, employs two-stage retrieval for long-range factual anchoring, and constructs explicit cross-modal evidence chains. Extensive experiments demonstrate that FRACTAM enhances mainstream models' performance in complex strategic tasks, improving hidden intent detection and inference while maintaining robust perceptual accuracy. Our dataset is available at https://naislab.cn/datasets/MISID.

Shufang Lin, Muyang Chen, Xiabing Zhou +3

The Chinese University of Hong Kong, Shenzhen Shenzhen, GuangDong, China · Capital Normal University Beijing, China

Multimodal Intent Recognition Multimodal Large Language Models Multimodal Reasoning Multimodal Model Evaluation

Apr 14, 2026

MODIX: A Training-Free Multimodal Information-Driven Positional Index Scaling for Vision-Language Models

Vision-Language Models (VLMs) have achieved remarkable progress in multimodal understanding, yet their positional encoding mechanisms remain suboptimal. Existing approaches uniformly assign positional indices to all tokens, overlooking variations in information density within and across modalities, which leads to inefficient attention allocation where redundant visual regions dominate while informative content is underrepresented. We identify positional granularity as an implicit resource and propose MODIX (Multimodal Information-Driven Positional IndeX Scaling), a training-free framework that dynamically adapts positional strides based on modality-specific contributions. MODIX jointly models intra-modal density via covariance-based entropy and inter-modal interaction via cross-modal alignment to derive unified scores, which rescale positional indices to allocate finer granularity to informative modalities while compressing redundant ones, without requiring any modification to model parameters or architecture. Experiments across diverse architectures and benchmarks demonstrate that MODIX consistently improves multimodal reasoning and adaptively reallocates attention according to task-dependent information distributions, suggesting that positional encoding should be treated as an adaptive resource in Transformers for multimodal sequence modeling.

Ruoxiang Huang, Zhen Yuan

Peking University, China

Vision-Language Models VLM Adaptation Multimodal Reasoning

Apr 14, 2026

Modality-Native Routing in Agent-to-Agent Networks: A Multimodal A2A Protocol Extension

Preserving multimodal signals across agent boundaries is necessary for accurate cross-modal reasoning, but it is not sufficient. We show that modality-native routing in Agent-to-Agent (A2A) networks improves task accuracy by 20 percentage points over text-bottleneck baselines, but only when the downstream reasoning agent can exploit the richer context that native routing preserves. An ablation replacing LLM-backed reasoning with keyword matching eliminates the accuracy gap entirely (36% vs. 36%), establishing a two-layer requirement: protocol-level routing must be paired with capable agent-level reasoning for the benefit to materialize. We present MMA2A, an architecture layer atop A2A that inspects Agent Card capability declarations to route voice, image, and text parts in their native modality. On CrossModal-CS, a controlled 50-task benchmark with the same LLM backend, same tasks, and only the routing path varying, MMA2A achieves 52% task completion accuracy versus 32% for the text-bottleneck baseline (95% bootstrap CI on

Δ

TCA: [8, 32] pp; McNemar's exact

p = 0.006

). Gains concentrate on vision-dependent tasks: product defect reports improve by +38.5 pp and visual troubleshooting by +16.7 pp. This accuracy gain comes at a

1.8\times

latency cost from native multimodal processing. These results suggest that routing is a first-order design variable in multi-agent systems, as it determines the information available for downstream reasoning.

Vasundra Srinivasan

AI Architect, Author—Data Engineering for Multimodal AI (O’Reilly), Stanford School of Engineering

Multi-Agent Communication Multimodal Model Routing Multimodal Agents Multimodal Reasoning

Apr 13, 2026

Collaborative Multi-Agent Scripts Generation for Enhancing Imperfect-Information Reasoning in Murder Mystery Games

Vision-language models (VLMs) have shown impressive capabilities in perceptual tasks, yet they degrade in complex multi-hop reasoning under multiplayer game settings with imperfect and deceptive information. In this paper, we study a representative multiplayer task, Murder Mystery Games, which require inferring hidden truths based on partial clues provided by roles with different intentions. To address this challenge, we propose a collaborative multi-agent framework for evaluating and synthesizing high-quality, role-driven multiplayer game scripts, enabling fine-grained interaction patterns tailored to character identities (i.e., murderer vs. innocent). Our system generates rich multimodal contexts, including character backstories, visual and textual clues, and multi-hop reasoning chains, through coordinated agent interactions. We design a two-stage agent-monitored training strategy to enhance the reasoning ability of VLMs: (1) chain-of-thought based fine-tuning on curated and synthetic datasets that model uncertainty and deception; (2) GRPO-based reinforcement learning with agent-monitored reward shaping, encouraging the model to develop character-specific reasoning behaviors and effective multimodal multi-hop inference. Extensive experiments demonstrate that our method significantly boosts the performance of VLMs in narrative reasoning, hidden fact extraction, and deception-resilient understanding. Our contributions offer a scalable solution for training and evaluating VLMs under uncertain, adversarial, and socially complex conditions, laying the groundwork for future benchmarks in multimodal multi-hop reasoning under imperfect information.

Keyang Zhong, Junlin Xie, Hefeng Wu +2

Sun Yat-sen University · Shenzhen Loop Area Institute · The Chinese University of Hong Kong, Shenzhen

VLM Reasoning Multimodal Reasoning Multi-Agent LLM Systems RL for Language Model Reasoning Large Language Model-Based Role-Play Simulation

Apr 13, 2026

CFMS: A Coarse-to-Fine Multimodal Synthesis Framework for Enhanced Tabular Reasoning

Reasoning over tabular data is a crucial capability for tasks like question answering and fact verification, as it requires models to comprehend both free-form questions and semi-structured tables. However, while methods like Chain-of-Thought (CoT) introduce reasoning chains, purely symbolic methodes are inherently limited by their blindness to holistic visual patterns. To address this, we propose the Coarse-to-Fine Multimodal Synthesis framework (CFMS), a novel two-stage paradigm that hierarchically decouples high-level visual perception from granular symbolic reasoning. In the Coarse Stage, CFMS leverages the Multimodal Large Language Models (MLLMs) to perform a one-time synthesis of a multi-perspective knowledge tuple. This tuple subsequently serves as a dynamic reasoning map to guide the fine stage, where a symbolic engine executes a targeted and efficient sequence of iterative operations over the table. Extensive experiments on the WikiTQ and TabFact benchmarks demonstrate that CFMS achieves competitive accuracy. The framework exhibits particular robustness when handling large tables and when instantiated with smaller backbone models, validating its effectiveness and generalizability.

Qixian Huang, Hongqiang Lin, Tong Fu +5

Table QA Multimodal Reasoning Symbolic Reasoning Automated Fact-Checking

Apr 12, 2026

When Meaning Isn't Literal: Exploring Idiomatic Meaning Across Languages and Modalities

Idiomatic reasoning, deeply intertwined with metaphor and culture, remains a blind spot for contemporary language models, whose progress skews toward surface-level lexical and semantic cues. For instance, the Bengali idiom \foreignlanguage{bengali}{\char"0986\char"0999\char"09CD\char"0997\char"09C1 \char"09B0 \char"09AB\char"09B2 \char"099F\char"0995} (angur fol tok,

grapes are sour''): it encodes denial-driven rationalization, yet naive models latch onto the literal fox-and-grape imagery. Addressing this oversight, we present

Mediom,'' a multilingual, multimodal idiom corpus of 3,533 Hindi, Bengali, and Thai idioms, each paired with gold-standard explanations, cross-lingual translations, and carefully aligned text--image representations. We benchmark both large language models (textual reasoning) and vision-language models (figurative disambiguation) on Mediom, exposing systematic failures in metaphor comprehension. To mitigate these gaps, we propose ``HIDE,'' a Hinting-based Idiom Explanation framework that leverages error-feedback retrieval and targeted diagnostic cues for iterative reasoning refinement. Collectively, Mediom and HIDE establish a rigorous test bed and methodology for culturally grounded, multimodal idiom understanding embedded with reasoning hints in next-generation AI systems.

Sarmistha Das, Shreyas Guha, Suvrayan Bandyopadhyay +3

Department of Computer Science and Engineering, Indian Institute of Technology Patna, Patna, India · School of Information Technology, King Mongkut’s Institute of Technology Ladkrabang, Bangkok, Thailand

Retrieval-Augmented Reasoning Figurative Language Understanding Cross-Lingual Reasoning in Language Models Multimodal Reasoning Cross-Cultural Language Model Evaluation

Apr 11, 2026

SVSR: A Self-Verification and Self-Rectification Paradigm for Multimodal Reasoning

Current multimodal models often suffer from shallow reasoning, leading to errors caused by incomplete or inconsistent thought processes. To address this limitation, we propose Self-Verification and Self-Rectification (SVSR), a unified framework that explicitly integrates self-verification and self-rectification into the model's reasoning pipeline, substantially improving robustness and reliability in complex visual understanding and multimodal reasoning tasks. SVSR is built on a novel three-stage training paradigm. First, we construct a high-quality unified preference dataset by refining reasoning traces from pre-trained vision-language models, incorporating both forward and backward reasoning to embed self-reflective signals. Second, we perform cold-start supervised fine-tuning on this dataset to learn structured, multi-step reasoning behaviors. Third, we apply a Semi-online Direct Preference Optimization (Semi-online DPO) process, continuously augmenting the training corpus with high-quality, model-generated reasoning traces filtered by a powerful teacher VLM. This pipeline enables the model to learn, elicit, and refine its ability to self-verify and self-rectify. Extensive experiments across diverse benchmarks demonstrate that SVSR improves reasoning accuracy and enables stronger generalization to unseen tasks and question types. Notably, once trained with explicit self-reflective reasoning, the model also exhibits improved implicit reasoning ability, outperforming strong baselines even when no explicit reasoning traces are provided. These results highlight the potential of SVSR for building more dependable, introspective, and cognitively aligned multimodal systems.

Zhe Qian, Nianbing Su, Zhonghua Wang +6

South China Agricultural University · Monash University · University of Glasgow +5

Multimodal Reasoning Vision-Language Models Multimodal Robustness

Apr 11, 2026

Cognitive Pivot Points and Visual Anchoring: Unveiling and Rectifying Hallucinations in Multimodal Reasoning Models

Multimodal Large Reasoning Models (MLRMs) have achieved remarkable strides in visual reasoning through test time compute scaling, yet long chain reasoning remains prone to hallucinations. We identify a concerning phenomenon termed the Reasoning Vision Truth Disconnect (RVTD): hallucinations are strongly correlated with cognitive bifurcation points that often exhibit high entropy states. We attribute this vulnerability to a breakdown in visual semantic anchoring, localized within the network's intermediate layers; specifically, during these high uncertainty transitions, the model fails to query visual evidence, reverting instead to language priors. Consequently, we advocate a shift from solely outcome level supervision to augmenting it with fine grained internal attention guidance. To this end, we propose V-STAR (Visual Structural Training with Attention Reinforcement), a lightweight, holistic training paradigm designed to internalize visually aware reasoning capabilities. Central to our approach is the Hierarchical Visual Attention Reward (HVAR), integrated within the GRPO framework. Upon detecting high entropy states, this mechanism dynamically incentivizes visual attention across critical intermediate layers, thereby anchoring the reasoning process back to the visual input. Furthermore, we introduce the Forced Reflection Mechanism (FRM), a trajectory editing strategy that disrupts cognitive inertia by triggering reflection around high entropy cognitive bifurcation points and encouraging verification of subsequent steps against the visual input, thereby translating external debiasing interventions into an intrinsic capability for hallucination mitigation.

Zhe Qian, Yanbiao Ma, Zhuohan Ouyang +7

South China Agricultural University · Gaoling School of Artificial Intelligence, Renmin University of China · South China Normal University +4

VLM Hallucination Mitigation Multimodal Reasoning Visually Grounded Reasoning RL for VLMs

Apr 10, 2026

GeoMMBench and GeoMMAgent: Toward Expert-Level Multimodal Intelligence in Geoscience and Remote Sensing

Recent advances in multimodal large language models (MLLMs) have accelerated progress in domain-oriented AI, yet their development in geoscience and remote sensing (RS) remains constrained by distinctive challenges: wide-ranging disciplinary knowledge, heterogeneous sensor modalities, and a fragmented spectrum of tasks. To bridge these gaps, we introduce GeoMMBench, a comprehensive multimodal question-answering benchmark covering diverse RS disciplines, sensors, and tasks, enabling broader and more rigorous evaluation than prior benchmarks. Using GeoMMBench, we assess 36 open-source and proprietary large language models, uncovering systematic deficiencies in domain knowledge, perceptual grounding, and reasoning--capabilities essential for expert-level geospatial interpretation. Beyond evaluation, we propose GeoMMAgent, a multi-agent framework that strategically integrates retrieval, perception, and reasoning through domain-specific RS models and tools. Extensive experimental results demonstrate that GeoMMAgent significantly outperforms standalone LLMs, underscoring the importance of tool-augmented agents for dynamically tackling complex geoscience and RS challenges.

Aoran Xiao, Shihao Cheng, Yonghao Xu +3

RIKEN AIP · Wuhan University · Linköping University +2

Remote Sensing Benchmarks Remote Sensing VQA Multimodal Agents Multimodal Reasoning Multi-Agent LLMs

Apr 9, 2026

Towards Responsible Multimodal Medical Reasoning via Context-Aligned Vision-Language Models

Medical vision-language models (VLMs) show strong performance on radiology tasks but often produce fluent yet weakly grounded conclusions due to over-reliance on a dominant modality. We introduce a context-aligned reasoning framework that enforces agreement across heterogeneous clinical evidence before generating diagnostic conclusions. The proposed approach augments a frozen VLM with structured contextual signals derived from radiomic statistics, explainability activations, and vocabulary-grounded semantic cues. Instead of producing free-form responses, the model generates structured outputs containing supporting evidence, uncertainty estimates, limitations, and safety notes. We observe that auxiliary signals alone provide limited benefit; performance gains emerge only when these signals are integrated through contextual verification. Experiments on chest X-ray datasets demonstrate that context alignment improves discriminative performance (AUC 0.918 to 0.925) while maintaining calibrated uncertainty. The framework also substantially reduces hallucinated keywords (1.14 to 0.25) and produces more concise reasoning explanations (19.4 to 15.3 words) without increasing model confidence (0.70 to 0.68). Cross-dataset evaluation on CheXpert further reveals that modality informativeness significantly influences reasoning behavior. These results suggest that enforcing multi-evidence agreement improves both reliability and trustworthiness in medical multimodal reasoning, while preserving the underlying model architecture.

Sumra Khan, Sagar Chhabriya, Aizan Zafar +5

Department of Computer Science, Salim Habib University, Karachi, Pakistan · Computer Science, Institute of Business Administration Sukkur, Pakistan · Center for Research in Computer Vision, University of Central Florida, USA +2

VLM Hallucination Mitigation Medical VLMs VLM Reasoning Radiology VLMs Vision-Language Models Multimodal Reasoning

Apr 9, 2026

Unified Multimodal Uncertain Inference

We introduce Unified Multimodal Uncertain Inference (UMUI), a multimodal inference task spanning text, audio, and video, where models must produce calibrated probability estimates of hypotheses conditioned on a premise in any modality or combination. While uncertain inference has been explored in text, extension to other modalities has been limited to single-modality binary entailment judgments, leaving no framework for fine-grained probabilistic reasoning in or across other modalities. To address this, we curate a human-annotated evaluation set with scalar probability judgments across audio, visual, and audiovisual settings, and additionally evaluate on existing text and audio benchmarks. We introduce CLUE (Calibrated Latent Uncertainty Estimation), which combines self-consistent teacher calibration and distribution-based confidence probing to produce calibrated predictions. We demonstrate that our 3B-parameter model achieves equivalent or stronger performance than zero-shot baselines up to 32B parameters across all modalities.

Dengjia Zhang, Alexander Martin, William Jurayj +3

Johns Hopkins University · Human Language Technology Center of Excellence

Probability Calibration Multimodal Reasoning Multimodal Model Evaluation Multimodal Understanding

Apr 9, 2026

Seeing but Not Thinking: Routing Distraction in Multimodal Mixture-of-Experts

Multimodal Mixture-of-Experts (MoE) models have achieved remarkable performance on vision-language tasks. However, we identify a puzzling phenomenon termed Seeing but Not Thinking: models accurately perceive image content yet fail in subsequent reasoning, while correctly solving identical problems presented as pure text. Through systematic analysis, we first verify that cross-modal semantic sharing exists in MoE architectures, ruling out semantic alignment failure as the sole explanation. We then reveal that visual experts and domain experts exhibit layer-wise separation, with image inputs inducing significant routing divergence from text inputs in middle layers where domain experts concentrate. Based on these findings, we propose the Routing Distraction hypothesis: when processing visual inputs, the routing mechanism fails to adequately activate task-relevant reasoning experts. To validate this hypothesis, we design a routing-guided intervention method that enhances domain expert activation. Experiments on three multimodal MoE models across six benchmarks demonstrate consistent improvements, with gains of up to 3.17% on complex visual reasoning tasks. Our analysis further reveals that domain expert identification locates cognitive functions rather than sample-specific solutions, enabling effective transfer across tasks with different information structures.

Haolei Xu, Haiwen Hong, Hongxing Li +7

Zhejiang University · Alibaba Group

VLM Reasoning Expert Routing Mixture-of-Experts Models Multimodal Reasoning

Apr 9, 2026

OpenVLThinkerV2: A Generalist Multimodal Reasoning Model for Multi-domain Visual Tasks

Group Relative Policy Optimization (GRPO) has emerged as the de facto Reinforcement Learning (RL) objective driving recent advancements in Multimodal Large Language Models. However, extending this success to open-source multimodal generalist models remains heavily constrained by two primary challenges: the extreme variance in reward topologies across diverse visual tasks, and the inherent difficulty of balancing fine-grained perception with multi-step reasoning capabilities. To address these issues, we introduce Gaussian GRPO (G

^2

RPO), a novel RL training objective that replaces standard linear scaling with non-linear distributional matching. By mathematically forcing the advantage distribution of any given task to strictly converge to a standard normal distribution,

\mathcal{N}(0,1)

, G

^2

RPO theoretically ensures inter-task gradient equity, mitigates vulnerabilities to heavy-tail outliers, and offers symmetric update for positive and negative rewards. Leveraging the enhanced training stability provided by G

^2

RPO, we introduce two task-level shaping mechanisms to seamlessly balance perception and reasoning. First, response length shaping dynamically elicits extended reasoning chains for complex queries while enforce direct outputs to bolster visual grounding. Second, entropy shaping tightly bounds the model's exploration zone, effectively preventing both entropy collapse and entropy explosion. Integrating these methodologies, we present OpenVLThinkerV2, a highly robust, general-purpose multimodal model. Extensive evaluations across 18 diverse benchmarks demonstrate its superior performance over strong open-source and leading proprietary frontier models.

Wenbo Hu, Xin Chen, Yan Gao-Tian +3

University of California, Los Angeles (UCLA)

Multimodal Large Language Models Multimodal Reasoning RL for VLMs Group Relative Policy Optimization

Apr 9, 2026

OmniJigsaw: Enhancing Omni-Modal Reasoning via Modality-Orchestrated Reordering

To extend the reinforcement learning post-training paradigm to omni-modal models for concurrently bolstering video-audio understanding and collaborative reasoning, we propose OmniJigsaw, a generic self-supervised framework built upon a temporal reordering proxy task. Centered on the chronological reconstruction of shuffled audio-visual clips, this paradigm strategically orchestrates visual and auditory signals to compel cross-modal integration through three distinct strategies: Joint Modality Integration, Sample-level Modality Selection, and Clip-level Modality Masking. Recognizing that the efficacy of such proxy tasks is fundamentally tied to puzzle quality, we design a two-stage coarse-to-fine data filtering pipeline, which facilitates the efficient adaptation of OmniJigsaw to massive unannotated omni-modal data. Our analysis reveals a ``bi-modal shortcut phenomenon'' in joint modality integration and demonstrates that fine-grained clip-level modality masking mitigates this issue while outperforming sample-level modality selection. Extensive evaluations on 15 benchmarks show substantial gains in video, audio, and collaborative reasoning, validating OmniJigsaw as a scalable paradigm for self-supervised omni-modal learning.

Yiduo Jia, Muzhi Zhu, Hao Zhong +7

Zhejiang University · Xiaomi Inc.

Audio-Visual Understanding Multimodal Reasoning Multimodal Pretraining Shortcut Learning

Apr 9, 2026

Multimodal Reasoning with LLM for Encrypted Traffic Interpretation: A Benchmark

Network traffic, as a key media format, is crucial for ensuring security and communications in modern internet infrastructure. While existing methods offer excellent performance, they face two key bottlenecks: (1) They fail to capture multidimensional semantics beyond unimodal sequence patterns. (2) Their black box property, i.e., providing only category labels, lacks an auditable reasoning process. We identify a key factor that existing network traffic datasets are primarily designed for classification and inherently lack rich semantic annotations, failing to generate human-readable evidence report. To address data scarcity, this paper proposes a Byte-Grounded Traffic Description (BGTD) benchmark for the first time, combining raw bytes with structured expert annotations. BGTD provides necessary behavioral features and verifiable chains of evidence for multimodal reasoning towards explainable encrypted traffic interpretation. Built upon BGTD, this paper proposes an end-to-end traffic-language representation framework (mmTraffic), a multimodal reasoning architecture bridging physical traffic encoding and semantic interpretation. In order to alleviate modality interference and generative hallucinations, mmTraffic adopts a jointly-optimized perception-cognition architecture. By incorporating a perception-centered traffic encoder and a cognition-centered LLM generator, mmTraffic achieves refined traffic interpretation with guaranteed category prediction. Extensive experiments demonstrate that mmTraffic autonomously generates high-fidelity, human-readable, and evidence-grounded traffic interpretation reports, while maintaining highly competitive classification accuracy comparing to specialized unimodal model (e.g., NetMamba). The source code is available at https://github.com/lgzhangzlg/Multimodal-Reasoning-with-LLM-for-Encrypted-Traffic-Interpretation-A-Benchmark

Longgang Zhang, Xiaowei Fu, Fuxiang Huang +1

School of Microelectronics and Communication Engineering, Chongqing University, Chongqing 400044, China. · School of Data Science, Lingnan University, Hong Kong, China.

Encrypted Network Traffic Analysis Encrypted Traffic Classification Multimodal Reasoning LLMs for Cybersecurity

Apr 9, 2026

Multimodal Latent Reasoning via Predictive Embeddings

Tool-augmented multimodal reasoning enables visual language models (VLMs) to improve perception by interacting with external tools (e.g., cropping, depth estimation). However, such approaches incur substantial inference overhead, require specialized supervision, and are prone to erroneous tool calls. We propose Pearl (Predictive Embedding Alignment for Reasoning in Latent space), a JEPA-inspired framework that learns from expert tool-use trajectories entirely in the latent space, eliminating the need for explicit tool invocation at inference time. Unlike reconstruction-based latent reasoning methods, which autoregressively generate latent tokens and suffer from training-inference mismatch and limited support for multi-step tool use, Pearl directly learns predictive embeddings from multimodal trajectories while preserving the standard vision-language generation pipeline: it is model-agnostic, simple to train, and naturally supports trajectories with multiple tool calls. Experiments across multiple perception benchmarks show that Pearl matches or outperforms standard supervised fine-tuning and reconstruction-based latent reasoning approaches. Furthermore, we provide empirical evidence that reconstruction-based methods primarily learn embeddings rather than image edits in latent space, motivating predictive embedding learning as a more principled alternative.

Ashutosh Adhikari, Mirella Lapata

School of Informatics University of Edimburgh

VLM Reasoning Multimodal Reasoning Latent Visual Reasoning Tool Use in VLMs Representation Learning

Apr 8, 2026

M-ArtAgent: Evidence-Based Multimodal Agent for Implicit Art Influence Discovery

Implicit artistic influence, although visually plausible, is often undocumented and thus poses a historically constrained attribution problem: resemblance is necessary but not sufficient evidence. Most prior systems reduce influence discovery to embedding similarity or label-driven graph completion, while recent multimodal large language models (LLMs) remain vulnerable to temporal inconsistency and unverified attributions. This paper introduces M-ArtAgent, an evidence-based multimodal agent that reframes implicit influence discovery as probabilistic adjudication. It follows a four-phase protocol consisting of Investigation, Corroboration, Falsification, and Verdict governed by a Reasoning and Acting (ReAct)-style controller that assembles verifiable evidence chains from images and biographies, enforces art-historical axioms, and subjects each hypothesis to adversarial falsification via a prompt-isolated critic. Two theory-grounded operators, StyleComparator for Wolfflin formal analysis and ConceptRetriever for ICONCLASS-based iconographic grounding, ensure that intermediate claims are formally auditable. On the balanced WikiArt Influence Benchmark-100 (WIB-100) of 100 artists and 2,000 directed pairs, M-ArtAgent achieves 83.7% positive-class F1, 0.666 Matthews correlation coefficient (MCC), and 0.910 area under the receiver operating characteristic curve (ROC-AUC), with leakage-control and robustness checks confirming that the gains persist when explicit influence phrases are masked. By coupling multimodal perception with domain-constrained falsification, M-ArtAgent demonstrates that implicit influence analysis benefits from historically grounded adjudication rather than pattern matching alone.

Hanyi Liu, Zhonghao Jiu, Minghao Wang +2

China Electronics Technology Group Co., Ltd., Beijing 100043, China · School of Information Science and Engineering, Southeast University, Nanjing 211189, China · Department of Chemical and Biological Engineering, Hong Kong University of Science and Technology, Hong Kong SAR, China +2

Multimodal Agents Multimodal Reasoning

Apr 8, 2026

What's Missing in Screen-to-Action? Towards a UI-in-the-Loop Paradigm for Multimodal GUI Reasoning

Existing Graphical User Interface (GUI) reasoning tasks remain challenging, particularly in UI understanding. Current methods typically rely on direct screen-based decision-making, which lacks interpretability and overlooks a comprehensive understanding of UI elements, ultimately leading to task failure. To enhance the understanding and interaction with UIs, we propose an innovative GUI reasoning paradigm called UI-in-the-Loop (UILoop). Our approach treats the GUI reasoning task as a cyclic Screen-UI elements-Action process. By enabling Multimodal Large Language Models (MLLMs) to explicitly learn the localization, semantic functions, and practical usage of key UI elements, UILoop achieves precise element discovery and performs interpretable reasoning. Furthermore, we introduce a more challenging UI Comprehension task centered on UI elements with three evaluation metrics. Correspondingly, we contribute a benchmark of 26K samples (UI Comprehension-Bench) to comprehensively evaluate existing methods' mastery of UI elements. Extensive experiments demonstrate that UILoop achieves state-of-the-art UI understanding performance while yielding superior results in GUI reasoning tasks.

Songze Li, Xiaoke Guo, Tianqi Liu +5

Zhejiang University · ZJU-Ant Group Joint Lab of Knowledge Graph

GUI Grounding Multimodal Reasoning Multimodal Model Evaluation

Apr 8, 2026

Walk the Talk: Bridging the Reasoning-Action Gap for Thinking with Images via Multimodal Agentic Policy Optimization

Recent advancements in Multimodal Large Language Models (MLLMs) have incentivized models to ``think with images'' by actively invoking visual tools during multi-turn reasoning. The common Reinforcement Learning (RL) practice of relying on outcome-based rewards ignores the fact that textual plausibility often masks executive failure, meaning that models may exhibit intuitive textual reasoning while executing imprecise or irrelevant visual actions within their agentic reasoning trajectories. This reasoning-action discrepancy introduces noise that accumulates throughout the multi-turn reasoning process, severely degrading the model's multimodal reasoning capabilities and potentially leading to training collapse. In this paper, we introduce Multimodal Agentic Policy Optimization (MAPO), bridging the gap between textual reasoning and visual actions generated by models within their Multimodal Chain-of-Thought (MCoT). Specifically, MAPO mandates the model to generate explicit textual descriptions for the visual content obtained via tool usage. We then employ a novel advantage estimation that couples the semantic alignment between these descriptions and the actual observations with the task reward. Theoretical findings are provided to justify the rationale behind MAPO, which inherently reduces the variance of gradients, and extensive experiments demonstrate that our method achieves superior performance across multiple visual reasoning benchmarks.

Wenhao Yang, Yu Xia, Jinlong Huang +10

National Key Laboratory for Novel Software Technology, Nanjing University, China · School of Artificial Intelligence, Nanjing University, China · AI Business, Alibaba Group +5

Multimodal Reasoning Multimodal Agents Multimodal CoT Reasoning

Apr 8, 2026

LAST: Leveraging Tools as Hints to Enhance Spatial Reasoning for Multimodal Large Language Models

Spatial reasoning is a cornerstone capability for intelligent systems to perceive and interact with the physical world. However, multimodal large language models (MLLMs) frequently suffer from hallucinations and imprecision when parsing complex geometric layouts. As data-driven scaling struggles to internalize structured geometric priors and spatial constraints, integrating mature, specialized vision models presents a compelling alternative. Despite its promise, applying this paradigm to spatial reasoning is hindered by two key challenges: The difficulty of invoking heterogeneous, parameter-rich tools, as well as the challenge of understanding and effectively leveraging their diverse low-level outputs (e.g., segmentation masks, depth maps) in high-level reasoning. To address these challenges, we propose LAST, a unified framework for tool-augmented spatial reasoning. LAST features an extensible interactive sandbox, termed LAST-Box, which abstracts heterogeneous tool invocations into atomic instructions and reusable spatial skills, returning multimodal hints (e.g., annotated images and textual descriptions) that can be directly consumed by LLMs. We further design a three-stage progressive training strategy that guides models from understanding tool outputs to proficient and adaptive tool invocation. Experiments on four datasets show that LAST-7B achieves around 20% performance gains over its backbone and outperforms strong proprietary closed-source LLMs, substantially enhancing reasoning on complex spatial tasks.

Shi-Yu Tian, Zhi Zhou, Kun-Yang Yu +5

National Key Laboratory for Novel Software Technology, Nanjing University, China

Visual Spatial Reasoning Multimodal Large Language Models Multimodal Reasoning Tool-Augmented Language Model Agents

Apr 8, 2026

RASR: Retrieval-Augmented Semantic Reasoning for Fake News Video Detection

Multimodal fake news video detection is a crucial research direction for maintaining the credibility of online information. Existing studies primarily verify content authenticity by constructing multimodal feature fusion representations or utilizing pre-trained language models to analyze video-text consistency. However, these methods still face the following limitations: (1) lacking cross-instance global semantic correlations, making it difficult to effectively utilize historical associative evidence to verify the current video; (2) semantic discrepancies across domains hinder the transfer of general knowledge, lacking the guidance of domain-specific expert knowledge. To this end, we propose a novel Retrieval-Augmented Semantic Reasoning (RASR) framework. First, a Cross-instance Semantic Parser and Retriever (CSPR) deconstructs the video into high-level semantic primitives and retrieves relevant associative evidence from a dynamic memory bank. Subsequently, a Domain-Guided Multimodal Reasoning (DGMP) module incorporates domain priors to drive an expert multimodal large language model in generating domain-aware, in-depth analysis reports. Finally, a Multi-View Feature Decoupling and Fusion (MVDFF) module integrates multi-dimensional features through an adaptive gating mechanism to achieve robust authenticity determination. Extensive experiments on the FakeSV and FakeTT datasets demonstrate that RASR significantly outperforms state-of-the-art baselines, achieves superior cross-domain generalization, and improves the overall detection accuracy by up to 0.93%.

Hui Li, Peien Ding, Jun Li +5

School of Informatics, Xiamen University, Xiamen, China · School of Computer Science and Information Security, Guilin University of Electronic Technology, Guilin, China · School of Computer and Big Data, Minjiang University, Fuzhou, China +2

Multimodal RAG Fake News Detection Multimodal Reasoning Multimodal Fusion

Apr 7, 2026

MTA-Agent: An Open Recipe for Multimodal Deep Search Agents

Multimodal large language models (MLLMs) have demonstrated strong capabilities in visual understanding, yet they remain limited in complex, multi-step reasoning that requires deep searching and integrating visual evidence with external knowledge. In this work, we address this challenge by constructing high-quality, verified multi-hop vision-language training data for multimodal deep-search agents. We propose a Multi-hop Tool-Augmented Agent for Evidence-based QA Synthesis (MTA-Agent), which automatically selects tools and their parameters to retrieve and validate evidence from both visual and textual sources and generates structured multi-hop question-answer trajectories. Starting from diverse VQA seed datasets, our pipeline produces a large-scale training dataset, MTA-Vision-DeepSearch, containing 21K high-quality multi-hop examples. The data is filtered through a multi-stage verification process to ensure factual consistency and answer uniqueness. Using MTA-Vision-DeepSearch, a 32B open-source multimodal search agent achieves state-of-the-art performance, reaching an average of 54.63% across six challenging benchmarks, outperforming GPT-5 (51.86%), Gemini-2.5-Pro (50.98%), and Gemini-3-Pro (54.46%) under the same tool settings. We further show that training on our data improves both reasoning depth and tool-use behavior, increasing the average number of steps from 2.27 to 4.28, and leading to more systematic and persistent search strategies. Additionally, we demonstrate that training can be performed without real-time tool calls by replaying cached interactions, significantly reducing training cost. Importantly, we present MTA-Agent as a fully open recipe for multimodal deep search: we release the entire dataset, training trajectories, and implementation details to enable reproducibility and future research on open multimodal search agents.

Xiangyu Peng, Can Qin, An Yan +4

Salesforce AI Research

Multimodal Search Agents Multimodal QA Multimodal Reasoning Multi-Hop QA Tool Use in VLMs

Apr 7, 2026

MMEmb-R1: Reasoning-Enhanced Multimodal Embedding with Pair-Aware Selection and Adaptive Control

MLLMs have been successfully applied to multimodal embedding tasks, yet their generative reasoning capabilities remain underutilized. Directly incorporating chain-of-thought reasoning into embedding learning introduces two fundamental challenges. First, structural misalignment between instance-level reasoning and pairwise contrastive supervision may lead to shortcut behavior, where the model merely learns the superficial format of reasoning. Second, reasoning is not universally beneficial for embedding tasks. Enforcing reasoning for all inputs may introduce unnecessary computation and latency, and can even obscure salient semantic signals for simple cases. To address these issues, we propose MMEmb-R1, an adaptive reasoning-based multimodal embedding framework. We formulate reasoning as a latent variable and introduce pair-aware reasoning selection that employs counterfactual intervention to identify reasoning paths beneficial for query-target alignment. Furthermore, we adopt reinforcement learning to selectively invoke reasoning only when necessary. Experiments on the MMEB-V2 benchmark demonstrate that our model achieves a score of 71.2 with only 4B parameters, establishing a new state-of-the-art while significantly reducing reasoning overhead and inference latency.

Yuchi Wang, Dingkang Yang, Haiyang Yu +5

MMLab, The Chinese University of Hong Kong · ByteDance Inc. · Fudan University

Multimodal Embedding Multimodal Reasoning Efficient Multimodal Inference

Apr 7, 2026

EpiBench: Benchmarking Multi-turn Research Workflows for Multimodal Agents

Scientific research follows multi-turn, multi-step workflows that require proactively searching the literature, consulting figures and tables, and integrating evidence across papers to align experimental settings and support reproducible conclusions. This joint capability is not systematically assessed in existing benchmarks, which largely under-evaluate proactive search, multi-evidence integration and sustained evidence use over time. In this work, we introduce EpiBench, an episodic multi-turn multimodal benchmark that instantiates short research workflows. Given a research task, agents must navigate across papers over multiple turns, align evidence from figures and tables, and use the accumulated evidence in the memory to answer objective questions that require cross paper comparisons and multi-figure integration. EpiBench introduces a process-level evaluation framework for fine-grained testing and diagnosis of research agents. Our experiments show that even the leading model achieves an accuracy of only 29.23% on the hard split, indicating substantial room for improvement in multi-turn, multi-evidence research workflows, providing an evaluation platform for verifiable and reproducible research agents.

Xuan Dong, Huanyang Zheng, Tianhao Niu +7

Harbin Institute of Technology

AI Agent Benchmarks AI Agents for Scientific Discovery Evidence Retrieval Multimodal Reasoning