mlsanityData to 7 Oct 2026

M2-Reasoning: Empowering MLLMs with Unified General and Spatial Reasoning

arXiv 2507.08306 ↗First posted 11 Jul 2025cs.AI cs.CL cs.CVcited by 16

On the map: LLM reasoning → Spatial and embodied reasoning

also in 1 more
Abstract (arXiv)

Recent advancements in Multimodal Large Language Models (MLLMs), particularly through Reinforcement Learning with Verifiable Rewards (RLVR), have significantly enhanced their reasoning abilities. However, a critical gap persists: these models struggle with dynamic spatial interactions, a capability essential for real-world applications. To bridge this gap, we introduce M2-Reasoning-7B, a model designed to excel in both general and spatial reasoning. Our approach integrates two key innovations: (1) a novel data pipeline that generates 294.2K high-quality data samples (168K for cold-start fine-tuning and 126.2K for RLVR), which feature logically coherent reasoning trajectories and have undergone comprehensive assessment; and (2) a dynamic multi-task training strategy with step-wise optimization to mitigate conflicts between data, and task-specific rewards for delivering tailored incentive signals. This combination of curated data and advanced training allows M2-Reasoning-7B to set a new state-of-the-art (SOTA) across 8 benchmarks, showcasing superior performance in both general and spatial reasoning domains.

Key facts. Cited by 16 papers in the map (data through 7 Oct 2026). Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning. Builds on 44 of its 56 references that are in the map. First posted on arXiv 11 Jul 2025.

Builds on

44 of 56 references.

References · 44

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2024The Llama 3 Herd of Modelscited by 14,192
  3. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
  4. 2025Qwen3 Technical Reportcited by 8,195
  5. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  6. 2025Qwen2.5-VL Technical Reportcited by 5,418
  7. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,669
  8. 2025InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Modelscited by 1,769
  9. 2023MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contextscited by 1,582
  10. 2024OpenAI o1 System Cardcited by 1,220
  11. 2025Kimi k1.5: Scaling Reinforcement Learning with LLMscited by 1,193
  12. 2023Gemini: A Family of Highly Capable Multimodal Modelscited by 810
32 more
  1. 2024Emu3: Next-Token Prediction is All You Needcited by 761
  2. 2024DeepSeek-V3 Technical Reportcited by 724
  3. 2025Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Modelscited by 653
  4. 2024LLaVA-NeXT-Interleave: Tackling Multi-image, Video, and 3D in Large Multimodal Modelscited by 571
  5. 2025LIMO: Less is More for Reasoningcited by 488
  6. 2025Video-R1: Reinforcing Video Reasoning in MLLMscited by 464
  7. 2025Kimi-VL Technical Reportcited by 381
  8. 2021Inter-GPS: Interpretable Geometry Problem Solving with Formal Language and Symbolic Reasoningcited by 376
  9. 2025Seed1.5-VL Technical Reportcited by 367
  10. 2025R1-Onevision: Advancing Generalized Multimodal Reasoning through Cross-Modal Formalizationcited by 364
  11. 2025MM-Eureka: Exploring the Frontiers of Multimodal Reasoning with Rule-based Reinforcement Learningcited by 294
  12. 2024We-Math: Does Your Large Multimodal Model Achieve Human-like Mathematical Reasoning?cited by 281
  13. 2025LMM-R1: Empowering 3B LMMs with Strong Reasoning Abilities Through Two-Stage Rule-Based RLcited by 244
  14. 2024LogicVista: Multimodal LLM Logical Reasoning Benchmark in Visual Contextscited by 240
  15. 2025SFT or RL? An Early Investigation into Training R1-Like Reasoning Large Vision-Language Modelscited by 209
  16. 2024Ovis: Structural Embedding Alignment for Multimodal Large Language Modelcited by 167
  17. 2025R1-Zero's "Aha Moment" in Visual Reasoning on a 2B Non-SFT Modelcited by 163
  18. 2025Seed1.5-Thinking: Advancing Superb Reasoning Models with Reinforcement Learningcited by 159
  19. 2025SoTA with Less: MCTS-Guided Sample Selection for Data-Efficient Visual Reasoning Self-Improvementcited by 144
  20. 2024DynaMath: A Dynamic Visual Benchmark for Evaluating Mathematical Reasoning Robustness of Vision Language Modelscited by 143
  21. 2025OpenVLThinker: Complex Vision-Language Reasoning via Iterative SFT-RL Cyclescited by 122
  22. 2024SAT: Dynamic Spatial Aptitude Training for Multimodal Language Modelscited by 111
  23. 2025MiMo-VL Technical Reportcited by 106
  24. 2025NoisyRollout: Reinforcing Visual Reasoning with Data Augmentationcited by 86
  25. 2025Visionary-R1: Mitigating Shortcuts in Visual Reasoning with Reinforcement Learningcited by 86
  26. 2025Unlocking Multimodal Mathematical Reasoning via Process Reward Modelcited by 64
  27. 2025Ming-Omni: A Unified Multimodal Model for Perception and Generationcited by 62
  28. 2025Scalable Vision Language Model Training via High Quality Data Curationcited by 54
  29. 2025Every FLOP Counts: Scaling a 300B Mixture-of-Experts LING LLM without Premium GPUscited by 22
  30. 2025WeThink: Toward General-purpose Vision-Language Reasoning via Reinforcement Learningcited by 21
  31. 2025M2-omni: Advancing Omni-MLLM for Comprehensive Modality Support with Competitive Performancecited by 19
  32. 2025Observe-R1: Unlocking Reasoning Abilities of MLLMs with Dynamic Progressive Reinforcement Learningcited by 13

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 10; 2026: 6.

16 papers.

Sort:

By area:

  1. 2025OpenMMReasoner: Pushing the Frontiers for Multimodal Reasoning with an Open and General Recipecited by 37
  2. 2025We-Math 2.0: A Versatile MathBook System for Incentivizing Visual Mathematical Reasoningcited by 32
  3. 2025Multimodal Spatial Reasoning in the Large Model Era: A Survey and Benchmarkscited by 31
  4. 2025V-Thinker: Interactive Thinking with Imagescited by 26
  5. 2025Ming-Flash-Omni: A Sparse, Unified Architecture for Multimodal Perception and Generationcited by 21
  6. 2025RynnEC: Bringing MLLMs into Embodied Worldcited by 11
  7. 2025Euclid's Gift: Enhancing Spatial Perception and Reasoning in Vision-Language Models via Geometric Surrogate Taskscited by 11
  8. 2026Thinking in Uncertainty: Mitigating Hallucinations in MLRMs with Latent Entropy-Aware Decodingcited by 11
  9. 2025COOPER: A Unified Model for Cooperative Perception and Reasoning in Spatial Intelligencecited by 9
  10. 2026Rethinking Token-Level Policy Optimization for Multimodal Chain-of-Thoughtcited by 7
  11. 2025CARE What Fails: Contrastive Anchored-REflection for Verifiable Multimodal Reasoningcited by 4
  12. 2026Think with Structured Grounding: Perceptual Reinforcement Learning for Chart and Visual-Tabular Understandingcited by 2
4 more
  1. 2025StaR-KVQA: Structured Reasoning Traces for Implicit-Knowledge Visual Question Answeringcited by 1
  2. 2026Dual Tuning for Reasoning Efficacy-Driven Data Curation in Multimodal LLM Trainingcited by 0
  3. 2026MHPR: Multidimensional Human Perception and Reasoning Benchmark for Large Vision-Languate Modelscited by 0
  4. 2026V-Rubrics: Visual Faithfulness via Rubric-Based Reinforcement Learningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Reinforced MLLM: A Survey on RL-Based Reasoning in Multimodal Large Language Models
  2. 2025Scaling and Beyond: Advancing Spatial Reasoning in MLLMs Requires New Recipes
  3. 2025SpaceR: Reinforcing MLLMs in Video Spatial Reasoning
  4. 2025Actial: Activate Spatial Reasoning Ability of Multimodal Large Language Models
  5. 2025ST-Think: How Multimodal Large Language Models Reason About 4D Worlds from Ego-Centric Videos
  6. 2025SVQA-R1: Reinforcing Spatial Reasoning in MLLMs via View-Consistent Reward Optimization
  7. 2024Exploring the Reasoning Abilities of Multimodal Large Language Models (MLLMs): A Comprehensive Survey on Emerging Trends in Multimodal Reasoning
  8. 2025Infi-MMR: Curriculum-based Unlocking Multimodal Reasoning via Phased Reinforcement Learning in Multimodal Small Language Models