mlsanityData to 13 Oct 2026

Attend to Evidence: Evidence-Anchored Spatial Attention Supervision for Multimodal RLVR

arXiv 2605.30912 ↗First posted 29 May 2026cs.CV cs.CLcited by 4

On the map: Multimodal LLMs / vision-language models → VLM benchmarks and visual reasoning

Abstract (arXiv)

Reinforcement learning with verifiable rewards (RLVR) improves vision-language models (VLMs) by optimizing outcome rewards derived from final answers. However, such outcome-only rewards do not tell the model which image regions justify an answer. For questions that require visual grounding, these rewards cannot distinguish responses supported by relevant visual evidence from those produced by language-prior shortcuts or lucky guesses. We introduce EASE (Evidence-Anchored Spatial Attention), which augments multimodal RLVR with visual-evidence process supervision. EASE converts annotated evidence regions into a smoothed visual-token target and uses it to guide response-to-image attention during RL training, but only on high-reward trajectories. The annotations are used solely as privileged training labels, while inference requires only the original image and question. Across Qwen2.5-VL-7B, Qwen3-VL-4B, and Qwen3-VL-8B, EASE raises average scores over DAPO by 2.5 to 3.1 points on perception, hallucination, visual math, and multimodal reasoning benchmarks. Diagnostics and ablations show that EASE better aligns visual attention with annotated evidence regions.

Key facts. Cited by 4 papers in the map (data through 13 Oct 2026). Placed in VLM benchmarks and visual reasoning, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 37 of its 48 references that are in the map. First posted on arXiv 29 May 2026.

Builds on

37 of 48 references.

Closest ancestors · 8

Same direction, newest first

  1. 2026Analyzing Reasoning Consistency in Large Multimodal Models under Cross-Modal Conflicts7 Jan
  2. 2025Visual Reasoning Tracer: Object-Level Grounded Reasoning Benchmark4 Dec
  3. 2025Bootstrapping Grounded Chain-of-Thought in Multimodal LLMs for Data-Efficient Model Adaptation3 Jul
  4. 2025Machine Mental Imagery: Empower Multimodal Reasoning with Latent Visual Tokens20 Jun
  5. 2024We-Math: Does Your Large Multimodal Model Achieve Human-like Mathematical Reasoning?1 Jul
  6. 2024MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?21 Mar
  7. 2023Ferret: Refer and Ground Anything Anywhere at Any Granularity11 Oct
  8. 2023MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts3 Oct

Other references · 29

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,091
  2. 2023Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detectioncited by 3,299
  3. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,729
  4. 2016CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoningcited by 1,450
  5. 2023Shikra: Unleashing Multimodal LLM's Referential Dialogue Magiccited by 898
  6. 2023Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decodingcited by 677
  7. 2025Visual-RFT: Visual Reinforcement Fine-Tuningcited by 506
  8. 2023OPERA: Alleviating Hallucination in Multi-Modal Large Language Models via Over-Trust Penalty and Retrospection-Allocationcited by 472
  9. 2023V*: Guided Visual Search as a Core Mechanism in Multimodal LLMscited by 470
  10. 2025R1-Onevision: Advancing Generalized Multimodal Reasoning through Cross-Modal Formalizationcited by 366
  11. 2025VL-Rethinker: Incentivizing Self-Reflection of Vision-Language Models with Reinforcement Learningcited by 320
  12. 2024LogicVista: Multimodal LLM Logical Reasoning Benchmark in Visual Contextscited by 242
17 more
  1. 2025SoTA with Less: MCTS-Guided Sample Selection for Data-Efficient Visual Reasoning Self-Improvementcited by 145
  2. 2024Devils in Middle Layers of Large Vision-Language Models: Interpreting, Detecting and Mitigating Object Hallucinations via Attention Lenscited by 141
  3. 2019Taking a HINT: Leveraging Explanations to Make Vision and Language Models More Groundedcited by 127
  4. 2016Human Attention in Visual Question Answering: Do Humans and Deep Networks Look at the Same Regions?cited by 111
  5. 2025Perception-Aware Policy Optimization for Multimodal Reasoningcited by 96
  6. 2022Super-CLEVR: A Virtual Benchmark to Diagnose Domain Robustness in Visual Reasoningcited by 90
  7. 2025NoisyRollout: Reinforcing Visual Reasoning with Data Augmentationcited by 86
  8. 2025More Thinking, Less Seeing? Assessing Amplified Hallucination in Multimodal Reasoning Modelscited by 85
  9. 2026MIRAGE: The Illusion of Visual Understandingcited by 61
  10. 2025Spotlight on Token Perception for Multimodal Reinforcement Learningcited by 38
  11. 2025Point-RFT: Improving Multimodal Reasoning with Visually Grounded Reinforcement Finetuningcited by 31
  12. 2025SpaCE-10: A Comprehensive Benchmark for Multimodal Large Language Models in Compositional Spatial Intelligencecited by 22
  13. 2025Perceptual-Evidence Anchored Reinforced Learning for Multimodal Reasoningcited by 20
  14. 2026Visually-Guided Policy Optimization for Multimodal Reasoningcited by 11
  15. 2025Chart-RVR: Reinforcement Learning with Verifiable Rewards for Explainable Chart Reasoningcited by 7
  16. 2026Large Vision-Language Models Get Lost in Attentioncited by 5
  17. 2026Chart-RL: Generalized Chart Comprehension via Reinforcement Learning with Verifiable Rewardscited by 3

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2026: 4.

4 papers.

  1. 2026CAVE: Competence-Aware Visual Boundary Evidence Alignment for Video Temporal Groundingcited by 1
  2. 2026VeriEvol: Scaling Multimodal Mathematical Reasoning via Verifiable Evol-Instructcited by 0
  3. 2026Industrial Anomaly Detection via Defect-Grounded Reasoning in Visual Latent Spacecited by 0
  4. 2026Look Where You Say You're Looking: Self-Grounded Attention for Visual Reasoningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Faithful-MR1: Faithful Multimodal Reasoning via Anchoring and Reinforcing Visual Attention
  2. 2026Same Reward, Different Skills: When Multimodal RL Learns to Look
  3. 2026See It, Say It, Sorted: An Iterative Training-Free Framework for Visually-Grounded Multimodal Reasoning in LVLMs
  4. 2025Seeing but Not Believing: Probing the Disconnect Between Visual Attention and Answer Correctness in VLMs
  5. 2026Improving Visual Reasoning with Iterative Evidence Refinement
  6. 2026Evidence-RL: Towards Evidence-intensive Visual Reasoning
  7. 2026What does RL improve for Visual Reasoning? A Frankenstein-Style Analysis
  8. 2026From Narrow to Panoramic Vision: Attention-Guided Cold-Start Reshapes Multimodal Reasoning