mlsanityData to 7 Oct 2026

Mind's Eye of LLMs: Visualization-of-Thought Elicits Spatial Reasoning in Large Language Models

arXiv 2404.03622 ↗First posted 4 Apr 2024cs.CLcited by 65

On the map: LLM reasoning → Spatial and embodied reasoning

Abstract (arXiv)

Large language models (LLMs) have exhibited impressive performance in language comprehension and various reasoning tasks. However, their abilities in spatial reasoning, a crucial aspect of human cognition, remain relatively unexplored. Human possess a remarkable ability to create mental images of unseen objects and actions through a process known as the Mind's Eye, enabling the imagination of the unseen world. Inspired by this cognitive capacity, we propose Visualization-of-Thought (VoT) prompting. VoT aims to elicit spatial reasoning of LLMs by visualizing their reasoning traces, thereby guiding subsequent reasoning steps. We employed VoT for multi-hop spatial reasoning tasks, including natural language navigation, visual navigation, and visual tiling in 2D grid worlds. Experimental results demonstrated that VoT significantly enhances the spatial reasoning abilities of LLMs. Notably, VoT outperformed existing multimodal large language models (MLLMs) in these tasks. While VoT works surprisingly well on LLMs, the ability to generate mental images to facilitate spatial reasoning resembles the mind's eye process, suggesting its potential viability in MLLMs. Please find the dataset and codes at https://microsoft.github.io/visualization-of-thought

Key facts. Cited by 65 papers in the map (data through 7 Oct 2026). Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning. Builds on 22 of its 50 references that are in the map. First posted on arXiv 4 Apr 2024.

Builds on

22 of 50 references.

Closest ancestors · 2

Same direction, newest first

  1. 2022Transfer Learning with Synthetic Corpora for Spatial Role Labeling and Reasoning30 Oct
  2. 2022StepGame: A New Benchmark for Robust Multi-Hop Spatial Reasoning in Texts18 Apr

Other references · 20

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  4. 2023Sparks of Artificial General Intelligence: Early experiments with GPT-4cited by 2,579
  5. 2023Large Language Modelscited by 774
  6. 2021ScanQA: 3D Question Answering for Spatial Scene Understandingcited by 340
  7. 2023Emergent Linear Representations in World Models of Self-Supervised Sequence Modelscited by 313
  8. 2015Towards AI-Complete Question Answering: A Set of Prerequisite Toy Taskscited by 309
  9. 2020Experience Grounds Languagecited by 241
  10. 2021Can Language Models Encode Perceptual Structure Without Grounding? A Case Study in Colorcited by 118
  11. 2020A Benchmark for Systematic Generalization in Grounded Language Understandingcited by 108
  12. 2023Evaluating Cognitive Maps and Planning in Large Language Models with CogEvalcited by 77
8 more
  1. 2021SpartQA: : A Textual Question Answering Benchmark for Spatial Reasoningcited by 75
  2. 2023Evaluating Spatial Understanding of Large Language Modelscited by 65
  3. 2023Coupling Large Language Models with Logic Programming for Robust and General Reasoning from Textcited by 59
  4. 2021A Systematic Investigation of Commonsense Knowledge in Large Language Modelscited by 56
  5. 2021Provable Limitations of Acquiring Meaning from Ungrounded Form: What Will Future Language Models Understand?cited by 55
  6. 2021Grounding Natural Language Instructions: Can Large Language Models Capture Spatial Information?cited by 10
  7. 2018Combining Deep Learning and Qualitative Spatial Reasoning to Learn Complex Structures from Sparse Examples with Noisecited by 4
  8. 2022A Symbolic Representation of Human Posture for Interpretable Learning and Reasoningcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 6; 2025: 38; 2026: 21.

65 papers.

Sort:

By area:

  1. 2025Multimodal Chain-of-Thought Reasoning: A Comprehensive Surveycited by 176
  2. 2025Visual Spatial Tuningcited by 85
  3. 2025Grounded Reinforcement Learning for Visual Reasoningcited by 80
  4. 2025Spatial Reasoning with Vision-Language Models in Ego-Centric Multi-View Scenescited by 62
  5. 2025Uni-cot: Towards Unified Chain-of-Thought Reasoning Across Text and Visioncited by 52
  6. 2025SSR: Enhancing Depth Perception in Vision-Language Models via Rationale-Guided Spatial Reasoningcited by 45
  7. 2025How Far are VLMs from Visual Spatial Intelligence? A Benchmark-Driven Perspectivecited by 41
  8. 2025A Survey of Test-Time Compute: From Intuitive Inference to Deliberate Reasoningcited by 40
  9. 2025VisRL: Intention-Driven Visual Perception via Reinforced Reasoningcited by 39
  10. 2025Fine-Grained Preference Optimization Improves Spatial Reasoning in VLMscited by 30
  11. 2024VideoINSTA: Zero-shot Long Video Understanding via Informative Spatial-Temporal Reasoning with LLMscited by 29
  12. 2024CoMT: A Novel Benchmark for Chain of Multi-modal Thought on Large Vision-Language Modelscited by 29
38 more
  1. 2025ViCrit: A Verifiable Reinforcement Learning Proxy Task for Visual Perception in VLMscited by 25
  2. 2025Human Cognitive Benchmarks Reveal Foundational Visual Gaps in MLLMscited by 12
  3. 2024Evaluating the Ability of Large Language Models to Reason about Cardinal Directionscited by 10
  4. 2025Enhancing Spatial Reasoning in Vision-Language Models via Chain-of-Thought Prompting and Reinforcement Learningcited by 9
  5. 2025Octopus: Agentic Multimodal Reasoning with Six-Capability Orchestrationcited by 8
  6. 2025CitySeeker: How Do VLMS Explore Embodied Urban Navigation With Implicit Human Needs?cited by 8
  7. 2026UniT: Unified Multimodal Chain-of-Thought Test-time Scalingcited by 8
  8. 2025PuzzleWorld: A Benchmark for Multimodal, Open-Ended Reasoning in Puzzlehuntscited by 7
  9. 2025Think in Games: Learning to Reason in Games via Reinforcement Learning with Large Language Modelscited by 7
  10. 2025CoFFT: Chain of Foresight-Focus Thought for Visual Language Modelscited by 6
  11. 2024Autonomous Imagination: Closed-Loop Decomposition of Visual-to-Textual Conversion in Visual Reasoning for Multimodal Large Language Modelscited by 5
  12. 2025Text Speaks Louder than Vision: ASCII Art Reveals Textual Biases in Vision-Language Modelscited by 5
  13. 2025AtomWorld: A Benchmark for Evaluating Spatial Reasoning in Large Language Models on Crystalline Materialscited by 5
  14. 2025EcoAlign: An Economically Rational Framework for Efficient LVLM Alignmentcited by 5
  15. 2024ASCIIEval: Benchmarking Models' Visual Perception in Text Strings via ASCII Artcited by 4
  16. 2025Probing LLM World Models: Enhancing Guesstimation with Wisdom of Crowds Decodingcited by 4
  17. 2025MathOPEval: A Fine-grained Evaluation Benchmark for Visual Operations of MLLMs in Mathematical Reasoningcited by 4
  18. 2025Thinking with Images via Self-Calling Agentcited by 4
  19. 2026Visual Latents Know More Than They Say: Unsilencing Latent Reasoning in MLLMscited by 4
  20. 2025Large Language and Reasoning Models are Shallow Disjunctive Reasonerscited by 3
  21. 2025GraphicBench: A Planning Benchmark for Graphic Design with Language Agentscited by 3
  22. 2024iToT: An Interactive System for Customized Tree-of-Thought Generationcited by 2
  23. 2025Navigating Motion Agents in Dynamic and Cluttered Environments through LLM Reasoningcited by 2
  24. 2025Empowering Multimodal LLMs with External Tools: A Comprehensive Surveycited by 2
  25. 2025Constructing coherent spatial memory in LLM agents through graph rectificationcited by 2
  26. 2025MACEval: A Multi-Agent Continual Evaluation Network for Large Modelscited by 2
  27. 2026Planning with Language and Generative Models: Toward General Reward-Guided Wireless Network Designcited by 2
  28. 2026STAR: Mitigating Cascading Errors in Spatial Reasoning via Turn-point Alignment and Segment-level DPOcited by 2
  29. 2025Can LLMs Learn to Map the World from Local Descriptions?cited by 1
  30. 2025LLM-Driven Self-Refinement for Embodied Drone Task Planningcited by 1
  31. 2025AuditoryBench++: Can Language Models Understand Auditory Knowledge without Hearing?cited by 1
  32. 2025Artificial Phantasia: Emergent Mental Imagery in Large Language Modelscited by 1
  33. 2025From Building Blocks to Planning: Multi-Step Spatial Reasoning in LLMs with Reinforcement Learningcited by 1
  34. 20263D-RFT: Reinforcement Fine-Tuning for Video-based 3D Scene Understandingcited by 1
  35. 2026Spectral-Progressive Thought Flow for Lightweight Multimodal Reasoningcited by 1
  36. 2026From Hallucination to Grounding: Diagnosing Visual Spatial Intelligence via CRISPcited by 1
  37. 2026GeoAnchor: Collaborative Reasoning via Latent Decomposition for 3D Spatial Understandingcited by 1
  38. 2026Beyond Visual CoT: Internalized Visual Thinking for Proactive Video Reasoningcited by 1

50 most cited of 65 All 65 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Imagine while Reasoning in Space: Multimodal Visualization-of-Thought
  2. 2025Imagine in Space: Exploring the Frontier of Spatial Intelligence and Reasoning Efficiency in Vision Language Models
  3. 2024Is A Picture Worth A Thousand Words? Delving Into Spatial Reasoning for Vision Language Models
  4. 2025Spatial Reasoning in Multimodal Large Language Models: A Survey of Tasks, Benchmarks and Methods
  5. 2025Reinforcing Spatial Reasoning in Vision-Language Models with Interwoven Thinking and Visual Drawing
  6. 202511Plus-Bench: Demystifying Multimodal LLM Spatial Reasoning with Cognitive-Inspired Analysis
  7. 2026Do LLMs Build Spatial World Models? Evidence from Grid-World Maze Tasks
  8. 2026Chain-of-Thought Degrades Visual Spatial Reasoning Capabilities of Multimodal LLMs