Self-Correction in VLMs

VLM: Vision-Language Model

Follow this topic Ask about this topic

Latest papers

17 papers2 papers in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.

MIRA: Reinforcing Multimodal Reasoning via Deceptive Contextual Augmentation
Nov 2, 2026Zhihan Yin, Jianxin Liang, Yifeng Yao +5Reasoning for VLMs VLM Robustness
Peking University · ByteDance · Wangxuan Institute of Computer Technology +1
Cycle-of-Thought: Self-Evaluative Iterative Reasoning for Table Visual Question Answering in Document Images
Oct 23, 2026Somraj Gautam, Gaurav HaritVisual Question Answering Reasoning for VLMs
IIT Jodhpur, Rajasthan, India.
GroundSight at GroundLM 2026 Shared Tasks: GoldenViewVQA
Oct 8, 2026Kun Wang, Yupeng Hu, Ruping Cao +4Self-Correction in VLMs Visual Question Answering
School of Software, Shandong University, Jinan, China · School of Computing, National University of Singapore, Singapore · School of Computer Science and Technology, Harbin Institute of Technology (Shenzhen), Shenzhen, China
When to Rethink: Learning Multi-Perspective Self-Verification for Vision-Language Models
Oct 4, 2026Ziquan Zhu, Hanruo Zhu, Si-Yuan Lu +10Self-Correction in VLMs VLM Reasoning
University of Leicester · Nanjing University of Posts and Telecommunications · University of North Carolina at Chapel Hill +5
LOCI: A Locator-Critic with Refinement Loop
Aug 31, 2026Walid Bousselham, Mathilde Caron, Arsha Nagrani +1Visual Question Answering Vision-Language Grounding
Google DeepMind
SAVER: Selective Auditing of Verbal Evidence for Error Recovery in VLM Change Reasoning
Aug 24, 2026Youdi LiSelf-Correction in VLMs VLM Reasoning
Panasonic Connect Co., Ltd. Tokyo, Japan
No Training, Better Flights: Test-Time Scaled VLMs for UAV Navigation
Jul 21, 2026Feinan Cheng, Dongliang Xu, Wenli Nong +4Vision-Language Models UAV Navigation
Shandong University Weihai, China · China University of Petroleum (East China) Qingdao, China · MBZUAI Abu Dhabi, United Arab Emirates
SVR-R1: Bootstrapping Multi-modal Reasoning with Self-verification in Reinforcement Learning
Jul 13, 2026Mingyuan Wu, Jingcheng Yang, Shengyi Qian +11Vision-Language Models Multimodal Reasoning
University of Illinois Urbana-Champaign · Meta
The Count Is There, but Misaligned: Understanding and Correcting Counting Failures in VLMs
Jul 10, 2026Ahmed Oumar El-Shangiti, Abzal Nurgazy, Hilal AlQuabeh +2Vision-Language Models VLM Interpretability
MBZUAI · Imperial College London
Visually Grounded Self-Reflection for Vision-Language Models via Reinforcement Learning
Jul 2, 2026Liyan Tang, Fangcong Yin, Greg DurrettVisually Grounded Reasoning Vision-Language Models
The University of Texas at Austin · New York University
ESC: Emotional Self-Correction for Reliable Vision-Language Models
Jul 1, 2026Tien-Huy Nguyen, Minh-Nhat Nguyen, Nguyen Nhat Huy +9LLM Self-Correction VLM Reasoning
GenAI4E Lab · University of Information Technology, Ho Chi Minh City, Vietnam · Vietnam National University, Ho Chi Minh City, Vietnam +10
Iterative Visual Thinking and the Self-Correction Mirage in VLM Grounding
Jun 11, 2026Animesh Tripathy, Aswanth KrishnanVLM Evaluation Vision-Language Grounding
QpiAI India Pvt. Ltd.
Aligning with Your Own Voice: Self-Corrected Preference Learning for Hallucination Mitigation in LVLMs
Apr 27, 2026Byeonggeuk Lim, JungMin Yun, Junehyoung Kwon +2Vision-Language Models Hallucination in Language Models
Graduate School of Advanced Imaging Sciences, Multimedia and Film, Chung-Ang University · Department of Artificial Intelligence, Chung-Ang University
When VLMs 'Fix' Students: Identifying and Penalizing Over-Correction in the Evaluation of Multi-line Handwritten Math OCR
Apr 1, 2026Jin Seong, Wencke Liermann, Minho Kim +2Handwritten Mathematical Expression Recognition Self-Correction in VLMs
Electronics and Telecommunications Research Institute, Republic of Korea
Self-Correction Inside the Model: Leveraging Layer Attention to Mitigate Hallucinations in Large Vision Language Models
Feb 28, 2026April FuCross-Layer Attention Self-Correction in VLMs
Independent Researcher
OCR-Agent: Agentic OCR with Capability and Memory Reflection
Feb 24, 2026Shimin Wen, Zeyu Zhang, Xingdou Bian +5Self-Correction in VLMs Vision-Language Models
Southwest Minzu University · AI Geeks
Bridging Modality Disconnect in Self-Reflection via Closed-Loop Visually Grounded Verification
Feb 21, 2026Haoyu Zhang, Yuwei Wu, Pengxiang Li +6Self-Correction in VLMs Vision-Language Models
Beijing Institute of Technology · Shenzhen MSU-BIT University

Nov 2, 2026J/K move · Enter open · S save

MIRA: Reinforcing Multimodal Reasoning via Deceptive Contextual Augmentation

Zhihan Yin, Jianxin Liang, Yifeng Yao, Nonghai Zhang+4

Peking University · ByteDance · Wangxuan Institute of Computer Technology+1

Reasoning for VLMs VLM Robustness Multimodal Robustness RL for VLMs Self-Correction in VLMs

Open paper AskPDF arXiv

While Reinforcement Learning with Verifiable Rewards (RLVR) has significantly advanced multimodal reasoning, existing frameworks suffer from a critical "perception-reasoning gap." Due to an over-reliance on seemingly relevant visual tokens, even minor perceptual perturbations can propagate through the reasoning process, leading to compounding hallucinations. To address this issue, we propose MIRA, a training framework that improves robustness to erroneous visual contexts while enhancing logical reasoning ability. By injecting filtered, deceptive visual contexts during training, we construct challenging "hard traps" that stress-test the model's robustness to misleading information. Furthermore, we introduce a group-wise reflection-triggering mechanism that enables the model to autonomously detect and rectify inconsistencies between visual cues and logical constraints, which is ultimately internalized as an intrinsic policy behavior. Importantly, MIRA does not rely on human annotations. Experimental results across multiple reasoning and perception benchmarks show that MIRA yields an average 7% improvement on in-domain benchmarks, with consistent improvements on vision-centric benchmarks. These findings highlight that active error recovery is essential for robust multimodal reasoning. Comprehensive ablation studies and analyses further provide insights into why MIRA is effective.