mlsanityData to 14 Oct 2026

MindJourney: Test-Time Scaling with World Models for Spatial Reasoning

arXiv 2507.12508 ↗First posted 16 Jul 2025cs.CV cs.AI cs.ROcited by 50

On the map: World models → World models for language agents

also in 1 more
Abstract (arXiv)

Spatial reasoning in 3D space is central to human cognition and indispensable for embodied tasks such as navigation and manipulation. However, state-of-the-art vision-language models (VLMs) struggle frequently with tasks as simple as anticipating how a scene will look after an egocentric motion: they perceive 2D images but lack an internal model of 3D dynamics. We therefore propose MindJourney, a test-time scaling framework that grants a VLM with this missing capability by coupling it to a controllable world model based on video diffusion. The VLM iteratively sketches a concise camera trajectory, while the world model synthesizes the corresponding view at each step. The VLM then reasons over this multi-view evidence gathered during the interactive exploration. Without any fine-tuning, our MindJourney achieves over an average 7.7% performance boost on the representative spatial reasoning benchmark SAT, showing that pairing VLMs with world models for test-time scaling offers a simple, plug-and-play route to robust 3D reasoning. Meanwhile, our method also improves upon the test-time inference VLMs trained through reinforcement learning, which demonstrates the potential of our method that utilizes world models for test-time scaling.

Key facts. Cited by 50 papers in the map (data through 14 Oct 2026). Placed in World models for language agents, one of 11 directions in World models. Builds on 58 of its 72 references that are in the map. First posted on arXiv 16 Jul 2025.

Builds on

58 of 72 references.

References · 58

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2023GPT-4 Technical Reportcited by 18,396
  4. 2023Visual Instruction Tuningcited by 7,869
  5. 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
  6. 2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,386
  7. 2022BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generationcited by 4,322
  8. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,234
  9. 2023InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuningcited by 2,876
  10. 2025Wan: Open and Advanced Large-Scale Video Generative Modelscited by 2,708
  11. 2023InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Taskscited by 2,565
  12. 2024CogVideoX: Text-to-Video Diffusion Models with An Expert Transformercited by 2,261
38 more
  1. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 1,986
  2. 2025InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Modelscited by 1,834
  3. 2024Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scalingcited by 1,599
  4. 2021Grounded Language-Image Pre-trainingcited by 1,156
  5. 2024MiniCPM-V: A GPT-4V Level MLLM on Your Phonecited by 1,079
  6. 2017AI2-THOR: An Interactive 3D Environment for Visual AIcited by 869
  7. 2024Genie: Generative Interactive Environmentscited by 747
  8. 2023CogVLM: Visual Expert for Pretrained Language Modelscited by 708
  9. 2024SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilitiescited by 706
  10. 2024Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spacescited by 611
  11. 2023DL3DV-10K: A Large-Scale Scene Dataset for Deep Learning-based 3D Visioncited by 537
  12. 2023Learning Universal Policies via Text-Guided Video Generationcited by 522
  13. 2024CameraCtrl: Enabling Camera Control for Text-to-Video Generationcited by 460
  14. 2023Learning Interactive Real-World Simulatorscited by 458
  15. 2023GLaMM: Pixel Grounding Large Multimodal Modelcited by 455
  16. 2024Vista: A Generalizable Driving World Model with High Fidelity and Versatile Controllabilitycited by 347
  17. 2021Light Field Networks: Neural Scene Representations with Single-Evaluation Renderingcited by 332
  18. 2024InternLM-XComposer2: Mastering Free-form Text-Image Composition and Comprehension in Vision-Language Large Modelcited by 326
  19. 2024SpatialRGPT: Grounded Spatial Reasoning in Vision Language Modelscited by 311
  20. 2022Visual Spatial Reasoningcited by 293
  21. 2025ReCamMaster: Camera-Controlled Generative Rendering from A Single Videocited by 287
  22. 2024Navigation World Modelscited by 272
  23. 2023What's "up" with vision-language models? Investigating their struggle with spatial reasoningcited by 254
  24. 2023Self-Evaluation Guided Beam Search for Reasoningcited by 196
  25. 2024RoboDreamer: Learning Compositional World Models for Robot Imaginationcited by 195
  26. 2025Stable Virtual Camera: Generative View Synthesis with Diffusion Modelscited by 182
  27. 2025History-Guided Video Diffusioncited by 171
  28. 2023Video Language Planningcited by 162
  29. 2024AlphaMath Almost Zero: Process Supervision without Processcited by 162
  30. 2023EmbodiedScan: A Holistic Multi-Modal 3D Perception Suite Towards Embodied AIcited by 147
  31. 2025From Flatland to Space: Teaching Vision-Language Models to Perceive and Reason in 3Dcited by 138
  32. 2024LLaVA-Critic: Learning to Evaluate Multimodal Modelscited by 134
  33. 2024BoNBoN Alignment for Large Language Models and the Sweetness of Best-of-n Samplingcited by 121
  34. 2021Pathdreamer: A World Model for Indoor Navigationcited by 118
  35. 2024SAT: Dynamic Spatial Aptitude Training for Multimodal Language Modelscited by 114
  36. 2025Aether: Geometric-Aware Unified World Modelingcited by 97
  37. 2023OVM, Outcome-supervised Value Models for Planning in Mathematical Reasoningcited by 85
  38. 2024Calibrated Self-Rewarding Vision Language Modelscited by 78

50 most cited of 58 All 58 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 11; 2026: 39.

50 papers.

Sort:

By area:

  1. 2025Cambrian-S: Towards Spatial Supersensing in Videocited by 134
  2. 2025A Comprehensive Survey on World Models for Embodied AIcited by 68
  3. 2025WoW: Towards a World omniscient World model Through Embodied Interactioncited by 67
  4. 2026Do World Action Models Generalize Better than VLAs? A Robustness Studycited by 36
  5. 2026Think3D: Thinking with Space for Spatial Reasoningcited by 24
  6. 2025Can World Simulators Reason? Gen-ViRe: A Generative Visual Reasoning Benchmarkcited by 19
  7. 2026When and How Much to Imagine: Adaptive Test-Time Scaling with World Models for Visual Spatial Reasoningcited by 14
  8. 2026Robotic Video World Models: A Survey of Applications, Research Challenges, Future Directionscited by 13
  9. 2025SpatialDreamer: Incentivizing Spatial Reasoning via Active Mental Imagerycited by 11
  10. 2026An Efficient and Multi-Modal Navigation System with One-Step World Modelcited by 9
  11. 2025Beyond Flatlands: Unlocking Spatial Intelligence by Decoupling 3D Reasoning from Numerical Regressioncited by 8
  12. 2025Abstract 3D Perception for Spatial Intelligence in Vision-Language Modelscited by 7
38 more
  1. 2026H-WM: Robotic Task and Motion Planning Guided by Hierarchical World Modelcited by 7
  2. 2025Toward Ambulatory Vision: Learning Visually-Grounded Active View Selectioncited by 6
  3. 2026Aligning Agentic World Models via Knowledgeable Experience Learningcited by 5
  4. 2026Allocentric Perceiver: Disentangling Allocentric Reasoning from Egocentric Visual Priors via Frame Instantiationcited by 5
  5. 2026See, Remember, Explore: A Benchmark and Baselines for Streaming Spatial Reasoningcited by 4
  6. 2026World2VLM: Distilling World Model Imagination into VLMs for Dynamic Spatial Reasoningcited by 4
  7. 2026Towards Spatial Supersensing in the Wildcited by 4
  8. 2025Video4Spatial: Towards Visuospatial Intelligence with Context-Guided Video Generationcited by 3
  9. 2025OpenView: Empowering MLLMs with Out-of-view VQAcited by 3
  10. 2026HIMM: Human-Inspired Long-Term Memory Modeling for Embodied Exploration and Question Answeringcited by 3
  11. 2026SPACENUM: Revisiting Spatial Numerical Understanding in VLMscited by 3
  12. 2026Pondering the Way: Spatial-perceiving World Action Model for Embodied Navigationcited by 3
  13. 2026What if? Emulative Simulation with World Models for Situated Reasoningcited by 2
  14. 2026WorldMAP: Bootstrapping Vision-Language Navigation Trajectory Prediction with Generative World Modelscited by 2
  15. 2026World Models Meet Language Models: On the Complementarity of Concrete and Abstract Reasoningcited by 2
  16. 2026Predicting Camera Pose from Perspective Descriptions for Spatial Reasoningcited by 1
  17. 2026From Where Things Are to What They Are For: Benchmarking Spatial-Functional Intelligence in Multimodal LLMscited by 1
  18. 2026MCP-Cosmos: World Model-Augmented Agents for Complex Task Execution in MCP Environmentscited by 1
  19. 20263D-Belief: Embodied Belief Inference via Generative 3D World Modelingcited by 1
  20. 2026GeoWorld-VLM: Geometry from World Models for Vision-Language Modelscited by 1
  21. 2026Skill-3D: Evolving Scene-Aware Skills for Agentic 3D Spatial Reasoningcited by 1
  22. 2026DeWorldSG: Depth-Aware 3D Semantic Scene Graph Generation via World-Model Priorscited by 1
  23. 2026GeoAnchor: Collaborative Reasoning via Latent Decomposition for 3D Spatial Understandingcited by 1
  24. 2025Probing the effectiveness of World Models for Spatial Reasoning through Test-time Scalingcited by 0
  25. 2026Token Warping Helps MLLMs Look from Nearby Viewpointscited by 0
  26. 2026Reasmory: 3D Reconstruction as Explicit Memory for VLMs Spatial Reasoningcited by 0
  27. 2026UniCanvas: A Diffusion-base Unified Model for Text-in-Image Joint Generationcited by 0
  28. 2026Test-Time Scaling in Multimodal Foundation Models: A Comprehensive Survey of Generation and Reasoningcited by 0
  29. 2026SpatialSV: Internalizing Interpretable 3D Spatial Awareness in MLLMs via Task-Oriented Visual Supervisioncited by 0
  30. 2026Geo3R: Mitigating Spatial Reasoning Hallucination in Multimodal Large Language Modelscited by 0
  31. 2026It's the Decoding Format, Not the Perturbation: Auditing Consistency-Based Selection for Vision-Language Test-Time Scalingcited by 0
  32. 2026DUET-DINO: Simultaneous Cross-View World Modeling for Latent Planning in Robot Manipulationcited by 0
  33. 2026Beyond Visual Quality: Evaluating Physical Consistency under Ego-Motion with EgoGenEvalcited by 0
  34. 2026Active Spatial Inspection for Effective and Efficient Embodied Explorationcited by 0
  35. 2026MA-WAM: Multi-Agent World-Action Model for Test-Time Planningcited by 0
  36. 2026WM-VLM: Probing Internal World Models for Interleaved Visual-Textual Reasoningcited by 0
  37. 2026WOVEN: Weaving Visual World Modeling into Multimodal LLMscited by 0
  38. 2026SpatialASM: Internalizing Active Mentalization for Spatial Reasoningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025MindCube: Spatial Mental Modeling from Limited Views
  2. 2026SpatialAct: Probing Spatial Reasoning-to-Action Capabilities of VLM Agents in 3D Scenes
  3. 2026VCN-Bench: A Video-Contextualized Navigation Benchmark for Spatial Reasoning over Prior Visual Experience
  4. 2025Reasoning Path and Latent State Analysis for Multi-view Visual Spatial Reasoning: A Cognitive Science Perspective
  5. 2025Imagine in Space: Exploring the Frontier of Spatial Intelligence and Reasoning Efficiency in Vision Language Models
  6. 2026SpaceCast-Bench: Evaluating Predictive Spatial Reasoning in Vision-Language Models
  7. 20264DThinker: Thinking with 4D Imagery for Dynamic Spatial Understanding
  8. 2026From Sight to Foresight: Predictive Spatial Reasoning in Vision-Language Models