Cover Image

When Seeing Is not Enough: Revealing the Limits of Active Reasoning in MLLMs

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Liu, Hongcheng, Wang, Pingjie, Wang, Yuhao, Ou, Siqu, Wang, Yanfeng, Wang, Yu
Format:	Preprint
Published:	2025
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

M2K-VDG: Model-Adaptive Multimodal Knowledge Anchor Enhanced Video-grounded Dialogue Generation
by: Liu, Hongcheng, et al.
Published: (2024)

Bridging the Dynamic Perception Gap: Training-Free Draft Chain-of-Thought for Dynamic Multimodal Spatial Reasoning
by: Ou, Siqu, et al.
Published: (2025)

Cross-Modal Coreference Alignment: Enabling Reliable Information Transfer in Omni-LLMs
by: Liu, Hongcheng, et al.
Published: (2026)

Drawing the Line: Enhancing Trustworthiness of MLLMs Through the Power of Refusal
by: Wang, Yuhao, et al.
Published: (2024)

Med-PMC: Medical Personalized Multi-modal Consultation with a Proactive Ask-First-Observe-Next Paradigm
by: Liu, Hongcheng, et al.
Published: (2024)

Selecting Auxiliary Data via Neural Tangent Kernels for Low-Resource Domains
by: Wang, Pingjie, et al.
Published: (2025)

VocalBench-DF: A Benchmark for Evaluating Speech LLM Robustness to Disfluency
by: Liu, Hongcheng, et al.
Published: (2025)

Automatic Interactive Evaluation for Large Language Models with State Aware Patient Simulator
by: Liao, Yusheng, et al.
Published: (2024)

MM-SAP: A Comprehensive Benchmark for Assessing Self-Awareness of Multimodal Large Language Models in Perception
by: Wang, Yuhao, et al.
Published: (2024)

Do MLLMs Really See It: Reinforcing Visual Attention in Multimodal LLMs
by: Ou, Siqu, et al.
Published: (2026)

Towards Omni-RAG: Comprehensive Retrieval-Augmented Generation for Large Language Models in Medical Applications
by: Chen, Zhe, et al.
Published: (2025)

Miner:Mining Intrinsic Mastery for Data-Efficient RL in Large Reasoning Models
by: Jiang, Shuyang, et al.
Published: (2026)

VocalBench-zh: Decomposing and Benchmarking the Speech Conversational Abilities in Mandarin Context
by: Liu, Heyang, et al.
Published: (2025)

Stop When Enough: Adaptive Early-Stopping for Chain-of-Thought Reasoning
by: Sun, Renliang, et al.
Published: (2025)

DSVD: Dynamic Self-Verify Decoding for Faithful Generation in Large Language Models
by: Guo, YiQiu, et al.
Published: (2025)

VocalBench: Benchmarking the Vocal Conversational Abilities for Speech Interaction Models
by: Liu, Heyang, et al.
Published: (2025)

Decoding Linguistic Representations of Human Brain
by: Wang, Yu, et al.
Published: (2024)

HeteroRAG: A Heterogeneous Retrieval-Augmented Generation Framework for Medical Vision Language Tasks
by: Chen, Zhe, et al.
Published: (2025)

Adaptive Deep Reasoning: Triggering Deep Thinking When Needed
by: Wang, Yunhao, et al.
Published: (2025)

What MLLMs Learn about When they Learn about Multimodal Reasoning
by: Chung, Jiwan, et al.
Published: (2025)

Outcome Accuracy is Not Enough: Aligning the Reasoning Process of Reward Models
by: Wang, Binghai, et al.
Published: (2026)

Agentic Active Omni-Modal Perception for Multi-Hop Audio-Visual Reasoning
by: Xu, Ke, et al.
Published: (2026)

Seeing from Another Perspective: Evaluating Multi-View Understanding in MLLMs
by: Yeh, Chun-Hsiao, et al.
Published: (2025)

CS3-Bench: Evaluating and Enhancing Speech-to-Speech LLMs for Mandarin-English Code-Switching
by: Liu, Heyang, et al.
Published: (2025)

M$^3$AV: A Multimodal, Multigenre, and Multipurpose Audio-Visual Academic Lecture Dataset
by: Chen, Zhe, et al.
Published: (2024)

VocalNet-MDM: Accelerating Streaming Speech LLM via Self-Distilled Masked Diffusion Modeling
by: Cheng, Ziyang, et al.
Published: (2026)

Beyond Unimodal Shortcuts: MLLMs as Cross-Modal Reasoners for Grounded Named Entity Recognition
by: Ma, Jinlong, et al.
Published: (2026)

Visual Room 2.0: Seeing is Not Understanding for MLLMs
by: Li, Haokun, et al.
Published: (2025)

When Is Thinking Enough? Early Exit via Sufficiency Assessment for Efficient Reasoning
by: Xiang, Yang, et al.
Published: (2026)

Human Cognitive Benchmarks Reveal Foundational Visual Gaps in MLLMs
by: Huang, Jen-Tse, et al.
Published: (2025)

Leveraging Diverse Modeling Contexts with Collaborating Learning for Neural Machine Translation
by: Liao, Yusheng, et al.
Published: (2024)

VocalNet-M2: Advancing Low-Latency Spoken Language Modeling via Integrated Multi-Codebook Tokenization and Multi-Token Prediction
by: Wang, Yuhao, et al.
Published: (2025)

When Is Enough Not Enough? Illusory Completion in Search Agents
by: Ko, Dayoon, et al.
Published: (2026)

DICE: Structured Reasoning in LLMs through SLM-Guided Chain-of-Thought Correction
by: Li, Yiqi, et al.
Published: (2025)

Seeing the Poem: Image-Semantic Detection of AI-Generated Modern Chinese Poetry with MLLMs
by: Wang, Shanshan, et al.
Published: (2026)

When Looking Is Not Enough: Visual Attention Structure Reveals Hallucination in MLLMs
by: Cao, Fanpu, et al.
Published: (2026)

When Less is Enough: Efficient Inference via Collaborative Reasoning
by: Chen, Yilei, et al.
Published: (2026)

Post-decoder Biasing for End-to-End Speech Recognition of Multi-turn Medical Interview
by: Liu, Heyang, et al.
Published: (2024)

MM-THEBench: Do Reasoning MLLMs Think Reasonably?
by: Huang, Zhidian, et al.
Published: (2026)

DRQA: Dynamic Reasoning Quota Allocation for Controlling Overthinking in Reasoning Large Language Models
by: Yan, Kaiwen, et al.
Published: (2025)