mlsanityData to 14 Oct 2026

MindCube: Spatial Mental Modeling from Limited Views

arXiv 2506.21458 ↗First posted 26 Jun 2025cs.AI cs.CL cs.CVcited by 95

On the map: Multimodal LLMs / vision-language models → Embodied and 3D vision-language models

Abstract (arXiv)

Can Vision-Language Models (VLMs) imagine the full scene from just a few views, like humans do? Humans form spatial mental models naturally, internal representations of unseen space, to reason about layout, perspective, and motion. Our MindCube benchmark with 21,154 questions across 3,268 images exposes this critical gap, where existing VLMs exhibit near-random performance. Using MindCube, we systematically evaluate how well VLMs build robust spatial mental models through representing positions (cognitive mapping), orientations (perspective-taking), and dynamics (mental simulation for "what-if" movements). We then explore three approaches to help approximate spatial mental models in VLMs, focusing on incorporating unseen intermediate views, natural language reasoning chains, and cognitive maps. The significant improvement comes from a synergistic approach, "map-then-reason", that jointly trains the model to first generate a cognitive map and then reason upon it. By training models to reason over these internal maps, we boosted accuracy from 37.8% to 57.8% (+20.0%). Adding reinforcement learning pushed performance even further to 61.3% (+23.5%). Our key insight is that such scaffolding of spatial mental models, actively constructing and utilizing internal structured spatial representations with flexible reasoning processes, significantly improves understanding of unobservable space.

Key facts. Cited by 95 papers in the map (data through 14 Oct 2026). Placed in Embodied and 3D vision-language models, one of 16 directions in Multimodal LLMs / vision-language models. First posted on arXiv 26 Jun 2025.

Builds on

0 of 0 references.

No references here yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 32; 2026: 63.

95 papers.

Sort:

By area:

  1. 2025Cambrian-S: Towards Spatial Supersensing in Videocited by 134
  2. 2025Visual Spatial Tuningcited by 89
  3. 2025Scaling Spatial Intelligence with Multimodal Foundation Modelscited by 82
  4. 2025Think with 3D: Geometric Imagination Grounded Spatial Reasoning from Limited Viewscited by 65
  5. 2025ThinkMorph: Emergent Properties in Multimodal Interleaved Chain-of-Thought Reasoningcited by 56
  6. 2025G2VLM: Geometry Grounded Vision Language Model with Unified 3D Reconstruction and Spatial Reasoningcited by 44
  7. 2025How Far are VLMs from Visual Spatial Intelligence? A Benchmark-Driven Perspectivecited by 43
  8. 2025SpatialBench: Benchmarking Multimodal Large Language Models for Spatial Cognitioncited by 35
  9. 2026STEP3-VL-10B Technical Reportcited by 32
  10. 2025Holistic Evaluation of Multimodal LLMs on Spatial Intelligencecited by 29
  11. 2025Explain Before You Answer: A Survey on Compositional Visual Reasoningcited by 28
  12. 2026Think3D: Thinking with Space for Spatial Reasoningcited by 24
38 more
  1. 2025DSI-Bench: A Benchmark for Dynamic Spatial Intelligencecited by 23
  2. 2025Spatial Reasoning in Multimodal Large Language Models: A Survey of Tasks, Benchmarks and Methodscited by 23
  3. 2025Multimodal RewardBench 2: Evaluating Omni Reward Models for Interleaved Text and Imagecited by 20
  4. 2026RAGEN-2: Reasoning Collapse in Agentic RLcited by 20
  5. 2026Visual Generation Unlocks Human-Like Reasoning through Multimodal World Modelscited by 19
  6. 2025SpatialThinker: Reinforcing Scene Graph-Grounded Spatial Reasoning via Dense Rewardscited by 16
  7. 2025SpinBench: Perspective and Rotation as a Lens on Spatial Reasoning in VLMscited by 14
  8. 2026pySpatial: Generating 3D Visual Programs for Zero-Shot Spatial Reasoningcited by 14
  9. 2025Euclid's Gift: Enhancing Spatial Perception and Reasoning in Vision-Language Models via Geometric Surrogate Taskscited by 13
  10. 2025SpatialTree: How Spatial Abilities Branch Out in MLLMscited by 12
  11. 2026Spatial-TTT: Streaming Visual-based Spatial Intelligence with Test-Time Trainingcited by 11
  12. 2026S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligencecited by 11
  13. 2025Can World Models Benefit VLMs for World Dynamics?cited by 9
  14. 2025Yo'City: Personalized and Boundless 3D Realistic City Scene Generation via Self-Critic Expansioncited by 9
  15. 2025MomaGraph: State-Aware Unified Scene Graphs with Vision-Language Model for Embodied Task Planningcited by 9
  16. 2025CitySeeker: How Do VLMS Explore Embodied Urban Navigation With Implicit Human Needs?cited by 8
  17. 2026Map2Thought: Explicit 3D Spatial Reasoning via Metric Cognitive Mapscited by 8
  18. 2025Mind Meets Space: Rethinking Agentic Spatial Intelligence from a Neuroscience-inspired Perspectivecited by 7
  19. 2026Generation Models Know Space: Unleashing Implicit 3D Priors for Scene Understandingcited by 7
  20. 2026Holi-Spatial: Evolving Video Streams into Holistic 3D Spatial Intelligencecited by 6
  21. 2026Faithful GRPO: Improving Visual Spatial Reasoning in Multimodal Language Models via Constrained Policy Optimizationcited by 6
  22. 2026Chain-of-Thought Degrades Visual Spatial Reasoning Capabilities of Multimodal LLMscited by 6
  23. 2025EmbRACE-3K: Embodied Reasoning and Action in Complex Environmentscited by 5
  24. 2026Boosting MLLM Spatial Reasoning with Geometrically Referenced 3D Scene Representationscited by 5
  25. 2026Thinking with Imagination: Agentic Visual Spatial Reasoning with World Simulatorscited by 5
  26. 2026Vesta: A Generalist Embodied Reasoning Modelcited by 5
  27. 2026ACE-Brain-0.5: A Unified Embodied Foundational Model for Physical Agentic AIcited by 5
  28. 2025Video2Layout: Recall and Reconstruct Metric-Grounded Cognitive Map for Spatial Reasoningcited by 4
  29. 20264DThinker: Thinking with 4D Imagery for Dynamic Spatial Understandingcited by 4
  30. 2026MindEdit-Bench: Benchmarking Object-Level Counterfactual Spatial Reasoning in VLMs from In-the-Wild Photoscited by 4
  31. 2026Towards Spatial Supersensing in the Wildcited by 4
  32. 2025Seeing through Imagination: Learning Scene Geometry via Implicit Spatial World Modelingcited by 3
  33. 2025Reasoning Path and Latent State Analysis for Multi-view Visual Spatial Reasoning: A Cognitive Science Perspectivecited by 3
  34. 2025Video4Spatial: Towards Visuospatial Intelligence with Context-Guided Video Generationcited by 3
  35. 2026From Correspondence to Actions: Human-Like Multi-Image Spatial Reasoning in Multi-modal Large Language Modelscited by 3
  36. 2026ARK: A Dual-Axis Multimodal Retrieval Benchmark along Reasoning and Knowledgecited by 3
  37. 2026VIEW2SPACE: Studying Multi-View Visual Reasoning from Sparse Observationscited by 3
  38. 2026HiSpatial: Taming Hierarchical 3D Spatial Understanding in Vision-Language Modelscited by 3

50 most cited of 95 All 95 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Mind the Gap: Benchmarking Spatial Reasoning in Vision-Language Models
  2. 2025Imagine in Space: Exploring the Frontier of Spatial Intelligence and Reasoning Efficiency in Vision Language Models
  3. 2026WM-VLM: Probing Internal World Models for Interleaved Visual-Textual Reasoning
  4. 2026SpatialASM: Internalizing Active Mentalization for Spatial Reasoning
  5. 2026SpatialCLI: Learning to Reason With Spatial Tools, Then Without Them
  6. 2026SpatiaLab: Can Vision-Language Models Perform Spatial Reasoning in the Wild?
  7. 2026CityCube: Benchmarking Cross-view Spatial Reasoning on Vision-Language Models in Urban Environments
  8. 2026KilometerVision: A New Frontier for Large-Scale Spatial Intelligence in VLMs