mlsanityData to 7 Oct 2026

Spatial and embodied reasoning

1,483 papers11 this weekMedium confidence in this grouping: mostly one topic, with blurry edgesMedium confidence in this grouping: mostly one topic, with blurry edges · About

Keywords: spatial, spatial reasoning, reasoning, embodied, vision-language

Key facts. Spatial and embodied reasoning: 1,483 papers on arXiv since 2022 (data through 7 Oct 2026). 267 of its 1,483 mapped papers (18%) cite “SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities” (2024, arXiv 2401.12168). 207 of its 1,483 mapped papers (14%) cite “Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces” (2024, arXiv 2412.14171). 67 new papers 17 Sep–7 Oct.

What’s new

17 Sep – 7 Oct 2026

67 new papers, 17 Sep – 7 Oct; 11 this week.

  1. This week · 11
  2. Relational Abstractions for Spatial Reasoning with Diffusion Models
  3. Flow-of-Thought: A Framework for Visual Reasoning
  4. Event-Aligned Visual Action Reasoning for World Action Models
  5. SpaTime: Streaming Vision-Language Models for Spatio-temporal Reasoning
  6. 2D Spatial Reasoning with Adaptive Neural Cellular Automata
  7. AssemState: Manual and Physical-State-Guided Reasoning for Zero-shot Furniture Assembly
  8. Event-Driven Proactive Robot Assistance through Vision-Language Reasoning
  9. ElasticFit: Fit-Aware 3D Object Insertion via VLM Reasoning and Generative Adaptation
  10. What Words Keep of a Place: Zero-Shot Language Reasoning for Cross-View Geo-Localization
  11. SpatialChain: A Benchmark for Auditing Spatial Reasoning Faithfulness in VLMs
All 67
  1. Rotated, but How Far? Diagnosing and Improving Object-Rotation Reasoning in VLMs
  2. Earlier · 56
  3. Render to Reason: Novel-View Semantic Prediction Improves Spatial Understanding in VLMs
  4. GeoBridge-VLA: Geometry-Aware Residual Adaptation for Vision-Language-Action Models
  5. PreAct-Nav: Agentic Reasoning Before Action for Urban Navigation also fits: Chain-of-thought reasoning for driving
  6. Agentic AI with Structured CoT for Enhancing AI's Spatial Intelligence: Visualization and Reasoning of Rotation
  7. From Sight to Foresight: Predictive Spatial Reasoning in Vision-Language Models
  8. CORNAV: Construction-Aware Reasoning for Robot Navigation on Active Worksites
  9. GeoLatent: Geometry-Guided Latent Structuring with Routed Optimization for 3D Reasoning
  10. UniWAM: Unified World-Action Model
  11. Task-Adaptive Grounded 3D-Programmers Using 2D VLMs
  12. From Reasoning Failures to Composable Video Spatial Intelligence
  13. PhysVista: Benchmarking Physical Intelligence in VLMs via a Perception-Reasoning-Assessment Loop also fits: Multimodal reasoning
  14. WorldAuditBench: Interactive 3D World Auditing with Multimodal Agents
  15. Network-based Spatial Context Retrieval for Open-weight LLMs: A Faithfulness Benchmark for Grounded Geographic Reasoning
  16. Soft Spatial Reasoning
  17. SpatialCORE: Confidence-Aware Grounded Spatial Reasoning in Large Vision--Language Models
  18. InsightMap: Structured Spatial Modeling for Embodied Multimodal Reasoning
  19. Spatial-OPSD: Self-Improving Spatial Reasoning via Label-Free Self-Distillation
  20. SphMind: Towards Robust, Training-Free VLM-based Spatial Reasoning with a 360 Camera
  21. DualManip: Agentic Dynamic Manipulation via Dual-Path Semantic Reasoning and Geometric Adaptation
  22. LeRF: Learning Reference Coordinate Frames for Perspective Taking Reasoning
  23. Reasoning with Neural Cellular Automata also fits: Test-time compute and efficient thinking
  24. Who Is Left of Whom? Tracing Spatial Evidence and Role Binding in Relative-Position Reasoning
  25. JRDB-AVR: An Active Visual Reasoning Benchmark for Embodied Agents in Real-World Environments
  26. WM-VLM: Probing Internal World Models for Interleaved Visual-Textual Reasoning
  27. VCN-Bench: A Video-Contextualized Navigation Benchmark for Spatial Reasoning over Prior Visual Experience
  28. V-Gym: Enhancing Agentic Visual Reasoning via Skill-Data Co-Evolution also fits: Multimodal reasoning
  29. Long Time No See: Benchmarking VLMs for Out-of-Sight Spatiotemporal Reasoning in Egocentric Videos
  30. Predictive Semantic Safety: From Visual Physical Reasoning to Safety-Critical Control also fits: Chain-of-thought reasoning for driving
  31. NavHarness: Towards Lifelong Embodied Navigation
  32. PainterBench: A Figural Divergent-Thinking Benchmark for Tool-Using Language Models also fits: Multimodal reasoning
  33. Geometric Encoding for Spatial Reasoning in Vision-Language Models
  34. SpatialSkill: Self-Evolving Skills for Cross-View Spatial Reasoning
  35. Test-Time Spatial Reasoning for Robot Manipulation Using Generative Real-to-Sim
  36. SpatialSpeak: QA-Native Reconstruction with Local and Global Context for Spatial Chain-of-Thought Reasoning
  37. Resolving State-Representation Mismatch: State-Space Visual Reasoning for Open-Loop VLA Planning
  38. USAI-Quant: A Quantitative Reasoning Benchmark for Vision-Language Models in Built Environments
  39. DepthEvidence: Unifying Metric Depth Prediction and Geometric Reasoning in Multimodal Language Models
  40. World Action Agent: Harnessing VLMs for Robot Manipulation via World Action Rehearsal
  41. Beyond Spatial Benchmarks: From Spatial Reasoning to Navigation
  42. Seeing Is Not Measuring: Tool-Augmented Metric Spatial Reasoning for Vision-Language Models
  43. EgoTSR++: Egocentric Spatiotemporal Reasoning for Task Progress Understanding
  44. OCC4M: Object-Centric 4D Memory for Spatiotemporal Reasoning in Long-Horizon Manipulation
  45. DrGait: Biomechanically Grounded Visual Reasoning for Interpretable Clinical Gait Analysis also fits: Multimodal reasoning
  46. Spatial and Semantic Reasoning for LLM-Driven Robot Navigation via MCP
  47. Metric-Bench: Exploring In-context Spatial Metric Reasoning in VLMs for Indoor Scenes
  48. Fysiverse-3D-Vision Technical Report: Generating Executable 3D Worlds from Images through Unified Spatial Reasoning
  49. INTCORT: Training-Free Spatial Reasoning Enhancement for Vision-Language Models via Input Transformations and Confidence Routing
  50. Cognitive Action Reasoning for Proactive Robots from Human-Centered Multimodal Observations
  51. CausalWM: Causal Chain-of-Thought Reasoning for Embodied World Model also fits: Causal reasoning in LLMs
  52. Spatial-Interactor: Learning Spatial Reasoning through Interaction with the Observable Physical World
  53. H-VLA: Hierarchical Vision-Language-Action Model with Key-Action Reasoning and Motion Planning in a Unified Action Space
  54. Adaptive World Memory 3D Foundation Model for Scalable 3D Mapping, Localization, and Rendering
  55. Hiding in Plain Sight: A Diffusion-based Mitigation of Geolocation Privacy Leakage in Vision-Language Models also fits: Multimodal reasoning
  56. VeriFuse: Bounded Vision-Language Arbitration and Reason-Guided Refinement for Cooperative 3D Perception also fits: Chain-of-thought reasoning for driving
  57. A Qualitative Model for Reasoning about Path and Support
  58. CitySTAR: Structured and Topology-Aware Reasoning for Open-Vocabulary Urban 3D Grounding

Foundations

Most cited by papers here

  1. SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities2024cited by 267 here
  2. Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces2024cited by 207 here
  3. SpatialRGPT: Grounded Spatial Reasoning in Vision Language Models2024cited by 172 here
  4. 3D-LLM: Injecting the 3D World into Large Language Models2023cited by 110 here
  5. Visual Spatial Reasoning2022cited by 81 here

Lineage

The papers this direction builds on, oldest first.

Each paper on the line cites the one before.

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 313 here
  2. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 65 here
  3. 2022Large Language Models are Zero-Shot Reasonerscited by 61 here
  4. 2022Inner Monologue: Embodied Reasoning through Planning with Language Modelscited by 81 here
  5. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 94 here
  6. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 64 here
  7. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 171 here
  8. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 176 here
16 more it builds on, not in this chain
  1. 2022Visual Spatial Reasoningcited by 81 here
  2. 2022SQA3D: Situated Question Answering in 3D Scenescited by 65 here
  3. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 105 here
  4. 20233D-LLM: Injecting the 3D World into Large Language Modelscited by 110 here
  5. 2023MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contextscited by 67 here
  6. 2023What's "up" with vision-language models? Investigating their struggle with spatial reasoningcited by 76 here
  7. 2024SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilitiescited by 267 here
  8. 2024SpatialRGPT: Grounded Spatial Reasoning in Vision Language Modelscited by 172 here
  9. 2024Robotic Control via Embodied Chain-of-Thought Reasoningcited by 63 here
  10. 2024Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scalingcited by 76 here
  11. 20243DSRBench: A Comprehensive 3D Spatial Reasoning Benchmarkcited by 69 here
  12. 2024Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spacescited by 207 here
  13. 2025SpaceR: Reinforcing MLLMs in Video Spatial Reasoningcited by 64 here
  14. 2025Qwen3 Technical Reportcited by 104 here
  15. 2025MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligencecited by 59 here
  16. 2025InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiencycited by 85 here

The 24 most cited of 2,054.

Browse

30 of 1,483, ranked 6–35

  1. 6Inner Monologue: Embodied Reasoning through Planning with Language Models202281
  2. 7What's "up" with vision-language models? Investigating their struggle with spatial reasoning202376
  3. 83DSRBench: A Comprehensive 3D Spatial Reasoning Benchmark202469
  4. 9SQA3D: Situated Question Answering in 3D Scenes202265
  5. 10SpaceR: Reinforcing MLLMs in Video Spatial Reasoning202564
  6. 11Robotic Control via Embodied Chain-of-Thought Reasoning202463
  7. 12MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence202559
  8. 13OmniSpatial: Towards Comprehensive Spatial Reasoning Benchmark for Vision Language Models202558
  9. 14CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models202555
  10. 15Is A Picture Worth A Thousand Words? Delving Into Spatial Reasoning for Vision Language Models202451
All 30
  1. 16Reinforcing Spatial Reasoning in Vision-Language Models with Interwoven Thinking and Visual Drawing202550
  2. 17ViewSpatial-Bench: Evaluating Multi-perspective Spatial Localization in Vision-Language Models202549
  3. 18LL3DA: Visual Interactive Instruction Tuning for Omni-3D Understanding, Reasoning, and Planning202348
  4. 19SpatialLadder: Progressive Training for Spatial Reasoning in Vision-Language Models202547
  5. 20RoboRefer: Towards Spatial Referring with Reasoning in Vision-Language Models for Robotics202540
  6. 21SpatialPIN: Enhancing Spatial Reasoning Capabilities of Vision-Language Models through Prompting and Interacting 3D Priors202440
  7. 22ReKep: Spatio-Temporal Reasoning of Relational Keypoint Constraints for Robotic Manipulation202440
  8. 23SpatialCoT: Advancing Spatial Reasoning through Coordinate Alignment and Chain-of-Thought for Embodied Task Planning202538
  9. 24NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language Models202335
  10. 25SpatialReasoner: Towards Explicit and Generalizable 3D Spatial Reasoning202534
  11. 26Scene-LLM: Extending Language Model for 3D Visual Understanding and Reasoning202434
  12. 27Mind's Eye of LLMs: Visualization-of-Thought Elicits Spatial Reasoning in Large Language Models202434
  13. 28Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success202532
  14. 29Why Is Spatial Reasoning Hard for VLMs? An Attention Mechanism Perspective on Focus Areas202532
  15. 30Mind the Gap: Benchmarking Spatial Reasoning in Vision-Language Models202532
  16. 31Perspective-Aware Reasoning in Vision-Language Models via Mental Imagery Simulation202532
  17. 32An Empirical Analysis on Spatial Reasoning Capabilities of Large Multimodal Models202432
  18. 33Cosmos-Reason1: From Physical Common Sense To Embodied Reasoning202530
  19. 34StepGame: A New Benchmark for Robust Multi-Hop Spatial Reasoning in Texts202229
  20. 35Reasoning Paths with Reference Objects Elicit Quantitative Spatial Reasoning in Large Vision-Language Models202428