mlsanityData to 7 Oct 2026

Trace: A Taxonomy-Guided Environment for Multidomain Visual Reasoning

arXiv 2607.19790 ↗First posted 22 Jul 2026cs.CVcited by 1

On the map: LLM reasoning → Multimodal reasoning

also in 1 more
Abstract (arXiv)

Reinforcement learning with verifiable rewards (RLVR) has substantially improved language-model reasoning, yet its extension to vision-language models remains constrained by the lack of training data that are simultaneously broad, exactly verifiable, and reproducible. We introduce Trace, a taxonomy-guided environment for multidomain visual reasoning. Trace factorizes task construction into a scene grammar and an executable task program, separating visual realization from answer computation. A shared semantic state determines the rendered image, prompt, typed answer, verifier state, and replayable instance trace. The resulting environment comprises 1,000 tasks over 277 scene grammars and 11 visual domains, with controlled semantic and visual variation. RLVR on 64,000 Trace instances improves the macro-average across 24 external benchmarks by 3.51 percentage points for Qwen2.5-VL-3B and 4.06 points for Qwen2.5-VL-7B, providing evidence that broad procedural training can transfer beyond the generated task distributions. Project page: https://maveryn.github.io/trace/.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Multimodal reasoning, one of 17 directions in LLM reasoning. Builds on 52 of its 57 references that are in the map. First posted on arXiv 22 Jul 2026.

Builds on

52 of 57 references.

Closest ancestors · 28

Same direction, newest first

  1. 2026WorldBench: A Challenging and Visually Diverse Multimodal Reasoning Benchmark4 Jun
  2. 2026TRON: Targeted Rule-Verifiable Online Environments for Visual Reasoning RL1 Jun
  3. 2026Vero: An Open RL Recipe for General Visual Reasoning6 Apr
  4. 2026DeepVision-103K: A Visually Diverse, Broad-Coverage, and Verifiable Mathematical Dataset for Multimodal Reasoning18 Feb
  5. 2025PuzzleCraft: Exploration-Aware Curriculum Learning for Puzzle-Based RLVR in VLMs16 Dec
  6. 2025SPHINX: A Synthetic Environment for Visual Perception and Reasoning25 Nov
  7. 2025Composition-Grounded Data Synthesis for Visual Reasoning16 Oct
  8. 2025ViCrit: A Verifiable Reinforcement Learning Proxy Task for Visual Perception in VLMs11 Jun
  9. 2025WeThink: Toward General-purpose Vision-Language Reasoning via Reinforcement Learning9 Jun
  10. 2025Truth in the Few: High-Value Data Selection for Efficient Multi-Modal Reasoning5 Jun
  11. 2025SynthRL: Scaling Visual Reasoning with Verifiable Data Synthesis2 Jun
  12. 2025VisualSphinx: Large-Scale Synthetic Vision Logic Puzzles for RL29 May
16 more
  1. 2025MME-Reasoning: A Comprehensive Benchmark for Logical Reasoning in MLLMs27 May
  2. 2025Game-RL: Synthesizing Multimodal Verifiable Game Data to Boost VLMs' General Reasoning20 May
  3. 2025VisualPuzzles: Decoupling Multimodal Reasoning Evaluation from Domain Knowledge14 Apr
  4. 2025SoTA with Less: MCTS-Guided Sample Selection for Data-Efficient Visual Reasoning Self-Improvement10 Apr
  5. 2025VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model10 Apr
  6. 2025Reason-RFT: Reinforcement Fine-Tuning for Visual Reasoning of Vision Language Models26 Mar
  7. 2025R1-Onevision: Advancing Generalized Multimodal Reasoning through Cross-Modal Formalization13 Mar
  8. 2025LMM-R1: Empowering 3B LMMs with Strong Reasoning Abilities Through Two-Stage Rule-Based RL10 Mar
  9. 2025MM-Eureka: Exploring the Frontiers of Multimodal Reasoning with Rule-based Reinforcement Learning10 Mar
  10. 2025Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Models9 Mar
  11. 2025Visual-RFT: Visual Reinforcement Fine-Tuning3 Mar
  12. 2024LogicVista: Multimodal LLM Logical Reasoning Benchmark in Visual Contexts6 Jul
  13. 2024We-Math: Does Your Large Multimodal Model Achieve Human-like Mathematical Reasoning?1 Jul
  14. 2024PuzzleVQA: Diagnosing Multimodal Reasoning Challenges of Language Models with Abstract Visual Patterns20 Mar
  15. 2024Measuring Multimodal Mathematical Reasoning with MATH-Vision Dataset22 Feb
  16. 2023MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts3 Oct

Other references · 24

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
  2. 2025Qwen2.5-VL Technical Reportcited by 5,418
  3. 2016CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoningcited by 1,445
  4. 2024Are We on the Right Way for Evaluating Large Vision-Language Models?cited by 893
  5. 2024Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMscited by 822
  6. 2024MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?cited by 573
  7. 2024BLINK: Multimodal Large Language Models Can See but Not Perceivecited by 529
  8. 2024MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmarkcited by 494
  9. 2024VLMEvalKit: An Open-Source Toolkit for Evaluating Large Multi-Modality Modelscited by 443
  10. 2025OpenThoughts: Data Recipes for Reasoning Modelscited by 287
  11. 2023Teaching CLIP to Count to Tencited by 194
  12. 2025REASONING GYM: Reasoning Environments for Reinforcement Learning with Verifiable Rewardscited by 121
12 more
  1. 2025ChartQAPro: A More Diverse and Challenging Benchmark for Chart Question Answeringcited by 105
  2. 2024EmbSpatial-Bench: Benchmarking Spatial Understanding for Embodied Tasks with Large Vision-Language Modelscited by 100
  3. 2024TableVQA-Bench: A Visual Question Answering Benchmark on Multiple Table Domainscited by 68
  4. 2025Traceable Evidence Enhanced Visual Grounded Reasoning: Evaluation and Methodologycited by 67
  5. 2025Enigmata: Scaling Logical Reasoning in Large Language Models with Synthetic Verifiable Puzzlescited by 53
  6. 2025PhyX: Does Your Model Have the "Wits" for Physical Reasoning?cited by 37
  7. 2025Jigsaw-R1: A Study of Rule-based Visual Reinforcement Learning with Jigsaw Puzzlescited by 36
  8. 2024EvoChart: A Benchmark and a Self-Training Approach Towards Real-World Chart Understandingcited by 32
  9. 2024Task Me Anythingcited by 29
  10. 2025MoDoMoDo: Multi-Domain Data Mixtures for Multimodal LLM Reinforcement Learningcited by 29
  11. 2025CountQA: How Well Do MLLMs Count in the Wild?cited by 17
  12. 2025SpatialViz-Bench: A Cognitively-Grounded Benchmark for Diagnosing Spatial Visualization in MLLMscited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026ReSight-SMC: Two-Stage Power Sampling via Island SMC with Visual Scoutscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Learning Only with Images: Visual Reinforcement Learning with Reasoning, Rendering, and Visual Feedback
  2. 2026Be Faithful When Response: Returning Fluent and Grounded Answers for Vision-Language Models Reinforcement Learning
  3. 2026Faithful-MR1: Faithful Multimodal Reasoning via Anchoring and Reinforcing Visual Attention
  4. 2025Self-Rewarding Vision-Language Model via Reasoning Decomposition
  5. 2025Activating Visual Context and Commonsense Reasoning through Masked Prediction in VLMs
  6. 2026VICO: Visual Environments Co-Evolving for Vision-Language Model Reasoning
  7. 2025RewardMap: Tackling Sparse Rewards in Fine-grained Visual Reasoning via Multi-Stage Reinforcement Learning
  8. 2026Visually-Guided Policy Optimization for Multimodal Reasoning