mlsanityData to 14 Oct 2026

Traceable Evidence Enhanced Visual Grounded Reasoning: Evaluation and Methodology

arXiv 2507.07999 ↗First posted 10 Jul 2025cs.CV cs.AI cs.CLcited by 69

Not on the map. Why?

Cited by 48 in Multimodal LLMs / vision-language models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Models like OpenAI-o3 pioneer visual grounded reasoning by dynamically referencing visual regions, just like human "thinking with images". However, no benchmark exists to evaluate these capabilities holistically. To bridge this gap, we propose TreeBench (Traceable Evidence Evaluation Benchmark), a diagnostic benchmark built on three principles: (1) focused visual perception of subtle targets in complex scenes, (2) traceable evidence via bounding box evaluation, and (3) second-order reasoning to test object interactions and spatial hierarchies beyond simple object localization. Prioritizing images with dense objects, we initially sample 1K high-quality images from SA-1B, and incorporate eight LMM experts to manually annotate questions, candidate options, and answers for each image. After three stages of quality control, TreeBench consists of 405 challenging visual question-answering pairs, even the most advanced models struggle with this benchmark, where none of them reach 60% accuracy, e.g., OpenAI-o3 scores only 54.87. Furthermore, we introduce TreeVGR (Traceable Evidence Enhanced Visual Grounded Reasoning), a training paradigm to supervise localization and reasoning jointly with reinforcement learning, enabling accurate localizations and explainable reasoning pathways. Initialized from Qwen2.5-VL-7B, it improves V* Bench (+16.8), MME-RealWorld (+12.6), and TreeBench (+13.4), proving traceability is key to advancing vision-grounded reasoning. The code is available at https://github.com/Haochen-Wang409/TreeVGR.

Key facts. Cited by 69 papers in the map (data through 14 Oct 2026). Builds on 62 of its 81 references that are in the map. First posted on arXiv 10 Jul 2025.

Builds on

62 of 81 references.

References · 62

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  3. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  4. 2023Segment Anythingcited by 8,234
  5. 2023Visual Instruction Tuningcited by 7,869
  6. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,137
  7. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  8. 2025Qwen2.5-VL Technical Reportcited by 5,587
  9. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 5,045
  10. 2024Qwen2.5 Technical Reportcited by 4,942
  11. 2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,386
  12. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,234
38 more
  1. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,810
  2. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,719
  3. 2023MMBench: Is Your Multi-modal Model an All-around Player?cited by 2,141
  4. 2023MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGIcited by 2,080
  5. 2024HybridFlow: A Flexible and Efficient RLHF Frameworkcited by 2,057
  6. 2025InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Modelscited by 1,834
  7. 2023Evaluating Object Hallucination in Large Vision-Language Modelscited by 1,729
  8. 2023MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contextscited by 1,614
  9. 2022ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoningcited by 1,479
  10. 2024LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Modelscited by 1,464
  11. 2020DocVQA: A Dataset for VQA on Document Imagescited by 1,317
  12. 2024How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suitescited by 1,050
  13. 2023SEED-Bench: Benchmarking Multimodal LLMs with Generative Comprehensioncited by 940
  14. 2024Are We on the Right Way for Evaluating Large Vision-Language Models?cited by 914
  15. 2024Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMscited by 838
  16. 2024Eyes Wide Shut? Exploring the Visual Shortcomings of Multimodal LLMscited by 735
  17. 2025Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Modelscited by 665
  18. 2023HallusionBench: An Advanced Diagnostic Suite for Entangled Language Hallucination and Visual Illusion in Large Vision-Language Modelscited by 583
  19. 2024DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understandingcited by 572
  20. 2025VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Modelcited by 526
  21. 2025Visual-RFT: Visual Reinforcement Fine-Tuningcited by 513
  22. 2023V*: Guided Visual Search as a Core Mechanism in Multimodal LLMscited by 482
  23. 2024VLMEvalKit: An Open-Source Toolkit for Evaluating Large Multi-Modality Modelscited by 455
  24. 2025DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learningcited by 410
  25. 2025Seed1.5-VL Technical Reportcited by 373
  26. 2024MME-RealWorld: Could Your Multimodal LLM Challenge High-Resolution Real-World Scenarios that are Difficult for Humans?cited by 292
  27. 2024Visual CoT: Advancing Multi-Modal Language Models with a Comprehensive Dataset and Benchmark for Chain-of-Thought Reasoningcited by 280
  28. 2025Pixel Reasoner: Incentivizing Pixel-Space Reasoning with Curiosity-Driven Reinforcement Learningcited by 278
  29. 2020Detection and Tracking Meet Drones Challengecited by 263
  30. 2024Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimizationcited by 239
  31. 2024MMT-Bench: A Comprehensive Multimodal Benchmark for Evaluating Large Vision-Language Models Towards Multitask AGIcited by 193
  32. 2024Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Modelscited by 161
  33. 2025MME-CoT: Benchmarking Chain-of-Thought in Large Multimodal Models for Reasoning Quality, Robustness, and Efficiencycited by 125
  34. 2025GRIT: Teaching MLLMs to Think with Imagescited by 125
  35. 2025UniVG-R1: Reasoning Guided Universal Visual Grounding with Reinforcement Learningcited by 84
  36. 2024Benchmarking and Improving Detail Image Captioncited by 81
  37. 2025VGR: Visual Grounded Reasoningcited by 73
  38. 2021LaTr: Layout-Aware Transformer for Scene-Text VQAcited by 72

50 most cited of 62 All 62 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 19; 2026: 50.

69 papers.

Sort:

By area:

  1. 2025A Survey of Context Engineering for Large Language Modelscited by 126
  2. 2025DeepEyesV2: Toward Agentic Multimodal Modelcited by 125
  3. 2025Thinking With Videos: Multimodal Tool-Augmented Reinforcement Learning for Long Video Reasoningcited by 91
  4. 2026Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillationcited by 77
  5. 2026Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perceptioncited by 50
  6. 2025Open-o3-Video: Grounded Video Reasoning with Explicit Spatio-Temporal Evidencecited by 49
  7. 2025Skywork-R1V4: Toward Agentic Multimodal Intelligence through Interleaved Thinking with Images and DeepResearchcited by 27
  8. 2026Embed-RL: Reinforcement Learning for Reasoning-Driven Multimodal Embeddingscited by 24
  9. 2025AdaTooler-V: Adaptive Tool-Use for Images and Videoscited by 21
  10. 2025Grasp Any Region: Towards Precise, Contextual Pixel Understanding for Multimodal LLMscited by 18
  11. 2025Think Twice to See More: Iterative Visual Reasoning in Medical VLMscited by 13
  12. 2025Scheduling Your LLM Reinforcement Learning with Reasoning Treescited by 12
38 more
  1. 2025Visual Reasoning Tracer: Object-Level Grounded Reasoning Benchmarkcited by 11
  2. 2025A Benchmark for Ultra-High-Resolution Remote Sensing MLLMscited by 11
  3. 2026SAMTok: Representing Any Mask with Two Wordscited by 11
  4. 2025DeFacto: Counterfactual Thinking with Images for Enforcing Evidence-Grounded and Faithful Reasoningcited by 10
  5. 2026VISD: Enhancing Video Reasoning via Structured Self-Distillationcited by 10
  6. 2026Thinking Without Images: Internalizing Visual Manipulation with On-Policy Self-Distillationcited by 9
  7. 2025Revisiting the Necessity of Lengthy Chain-of-Thought in Vision-centric Reasoning Generalizationcited by 8
  8. 2025VG-Refiner: Towards Tool-Refined Referring Grounded Reasoning via Agentic Reinforcement Learningcited by 8
  9. 2026VideoZeroBench: Probing the Limits of Video MLLMs with Spatio-Temporal Evidence Verificationcited by 7
  10. 2026DeepScan: A Training-Free Framework for Visually Grounded Reasoning in Large Vision-Language Modelscited by 6
  11. 2026Faithful GRPO: Improving Visual Spatial Reasoning in Multimodal Language Models via Constrained Policy Optimizationcited by 6
  12. 2026Chain-of-Thought Degrades Visual Spatial Reasoning Capabilities of Multimodal LLMscited by 6
  13. 2025UI-UG: A Unified MLLM for UI Understanding and Generationcited by 5
  14. 2025Know-Show: Benchmarking Video-Language Models on Spatio-Temporal Grounded Reasoningcited by 5
  15. 2026Reliable Thinking with Imagescited by 5
  16. 2026VisionCoach: Reinforcing Grounded Video Reasoning via Visual-Perception Promptingcited by 4
  17. 2026Motion-o: Trajectory-Grounded Video Reasoningcited by 4
  18. 2026VISTAQA: Benchmarking Joint Visual Question Answering and Pixel-Level Evidencecited by 4
  19. 2026AVIS: Adaptive Test-Time Scaling for Vision-Language Modelscited by 4
  20. 2026PerceptionBench: Evaluating Atomic Visual Perception in Multimodal Large Language Modelscited by 4
  21. 2026HART: High-Resolution Annotation-Free Reasoning Technique through a Closed-loop Frameworkcited by 3
  22. 2025Guiding the Inner Eye: A Framework for Hierarchical and Flexible Visual Grounded Reasoningcited by 2
  23. 2026Seek-and-Solve: Benchmarking MLLMs for Visual Clue-Driven Reasoning in Daily Scenarioscited by 2
  24. 2026Are Tools Always Beneficial? Learning to Invoke Tools Adaptively for Dual-Mode Multimodal LLM Reasoningcited by 2
  25. 2026PixelEyes: Decoupling Perception and Reasoning for Pinpoint Visual Evidence Seekingcited by 2
  26. 2025SVAC: Scaling Is All You Need For Referring Video Object Segmentationcited by 1
  27. 2026Chatting with Images for Introspective Visual Thinkingcited by 1
  28. 2026See It, Say It, Sorted: An Iterative Training-Free Framework for Visually-Grounded Multimodal Reasoning in LVLMscited by 1
  29. 2026Test-time Scaling over Perception: Resolving the Grounding Paradox in Thinking with Imagescited by 1
  30. 2026CVSearch: Empowering Multimodal LLMs with Cognitive Visual Search for High-Resolution Image Perceptioncited by 1
  31. 2026DeepLatent: Think with Images via Parallel Latent Visual Reasoningcited by 1
  32. 2026Actor as Its Own Critic: Unifying Region Understanding and Localization via CycleGRPOcited by 1
  33. 2026Trace: A Taxonomy-Guided Environment for Multidomain Visual Reasoningcited by 1
  34. 2025Logo-VGR: Visual Grounded Reasoning for Open-world Logo Recognitioncited by 0
  35. 2026MARINER: A 3E-Driven Benchmark for Fine-Grained Perception and Complex Reasoning in Open-Water Environmentscited by 0
  36. 2026DRAGON: A Benchmark for Evidence-Grounded Visual Reasoning over Diagramscited by 0
  37. 2026Perceptual Flow Network for Visually Grounded Reasoningcited by 0
  38. 2026Starve to Perceive: Taming Lazy Perception in VLMs with Constrained Visual Bandwidthcited by 0

50 most cited of 69 All 69 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Beyond Accuracy: Evaluating Grounded Visual Evidence in Thinking with Images
  2. 2026VG-CoT: Towards Trustworthy Visual Reasoning via Grounded Chain-of-Thought
  3. 2025VER-Bench: Evaluating MLLMs on Reasoning with Fine-Grained Visual Evidence
  4. 2026Evidence-RL: Towards Evidence-intensive Visual Reasoning
  5. 2026Thinking with Visual Grounding
  6. 2026TACIT Benchmark: A Programmatic Visual Reasoning Benchmark for Generative and Discriminative Models
  7. 2026UltraVR: A Diagnostic Ultra-Resolution Image-VQA Benchmark for Evidence-Grounded Reasoning
  8. 2026Frame Evidence:A Benchmark for Reference Frames and Visual Evidence in Multi-Image Spatial Reasoning