mlsanityData to 14 Oct 2026

SpatialReasoner: Towards Explicit and Generalizable 3D Spatial Reasoning

arXiv 2504.20024 ↗First posted 28 Apr 2025cs.CVcited by 68

On the map: LLM reasoning → Spatial and embodied reasoning

Abstract (arXiv)

Despite recent advances on multi-modal models, 3D spatial reasoning remains a challenging task for state-of-the-art open-source and proprietary models. Recent studies explore data-driven approaches and achieve enhanced spatial reasoning performance by fine-tuning models on 3D-related visual question-answering data. However, these methods typically perform spatial reasoning in an implicit manner and often fail on questions that are trivial to humans, even with long chain-of-thought reasoning. In this work, we introduce SpatialReasoner, a novel large vision-language model (LVLM) that addresses 3D spatial reasoning with explicit 3D representations shared between multiple stages--3D perception, computation, and reasoning. Explicit 3D representations provide a coherent interface that supports advanced 3D spatial reasoning and improves the generalization ability to novel question types. Furthermore, by analyzing the explicit 3D representations in multi-step reasoning traces of SpatialReasoner, we study the factual errors and identify key shortcomings of current LVLMs. Results show that our SpatialReasoner achieves improved performance on a variety of spatial reasoning benchmarks, outperforming Gemini 2.0 by 9.2% on 3DSRBench, and generalizes better when evaluating on novel 3D spatial reasoning questions. Our study bridges the 3D parsing capabilities of prior visual foundation models with the powerful reasoning abilities of large language models, opening new directions for 3D spatial reasoning.

Key facts. Cited by 68 papers in the map (data through 14 Oct 2026). Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning. Builds on 40 of its 58 references that are in the map. First posted on arXiv 28 Apr 2025.

Builds on

40 of 58 references.

Closest ancestors · 5

Same direction, newest first

  1. 2024Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces18 Dec
  2. 20243DSRBench: A Comprehensive 3D Spatial Reasoning Benchmark10 Dec
  3. 2024ReKep: Spatio-Temporal Reasoning of Relational Keypoint Constraints for Robotic Manipulation3 Sep
  4. 2024SpatialRGPT: Grounded Spatial Reasoning in Vision Language Models3 Jun
  5. 2024SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities22 Jan

Other references · 35

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  3. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  4. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  5. 2023Visual Instruction Tuningcited by 7,869
  6. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  7. 2025Qwen2.5-VL Technical Reportcited by 5,587
  8. 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
  9. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,780
  10. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,720
  11. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,211
  12. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 3,050
23 more
  1. 2024SAM 2: Segment Anything in Images and Videoscited by 2,863
  2. 2023MMBench: Is Your Multi-modal Model an All-around Player?cited by 2,141
  3. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 1,986
  4. 2016CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoningcited by 1,457
  5. 2024Depth Anything V2cited by 1,422
  6. 2025Understanding R1-Zero-Like Training: A Critical Perspectivecited by 1,345
  7. 2025s1: Simple test-time scalingcited by 1,341
  8. 2024Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMscited by 838
  9. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 672
  10. 2019The Neuro-Symbolic Concept Learner: Interpreting Scenes, Words, and Sentences From Natural Supervisioncited by 372
  11. 2024NaVILA: Legged Robot Vision-Language-Action Model for Navigationcited by 211
  12. 2023Physically Grounded Vision-Language Models for Robotic Manipulationcited by 165
  13. 2022Omni3D: A Large Benchmark and Model for 3D Object Detection in the Wildcited by 160
  14. 2024SpatialBot: Precise Spatial Understanding with Vision Language Modelscited by 152
  15. 2025LLM Post-Training: A Deep Dive into Reasoning Large Language Modelscited by 135
  16. 2024SAT: Dynamic Spatial Aptitude Training for Multimodal Language Modelscited by 114
  17. 2024Fast Best-of-N Decoding via Speculative Rejectioncited by 112
  18. 2022Super-CLEVR: A Virtual Benchmark to Diagnose Domain Robustness in Visual Reasoningcited by 92
  19. 2025A Survey on Post-training of Large Language Modelscited by 54
  20. 2024ImageNet3D: Towards General-Purpose Object-Level 3D Understandingcited by 25
  21. 2022Robust Category-Level 6D Pose Estimation with Coarse-to-Fine Rendering of Neural Featurescited by 23
  22. 2024Compositional 4D Dynamic Scenes Understanding with Physics Priors for Video Question Answeringcited by 18
  23. 20233D-Aware Visual Question Answering about Parts, Poses and Occlusionscited by 16

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 21; 2026: 47.

68 papers.

Sort:

By area:

  1. 2025The Landscape of Agentic Reinforcement Learning for LLMs: A Surveycited by 205
  2. 2025Visual Spatial Tuningcited by 89
  3. 2025How Far are VLMs from Visual Spatial Intelligence? A Benchmark-Driven Perspectivecited by 43
  4. 2025Fine-Grained Preference Optimization Improves Spatial Reasoning in VLMscited by 32
  5. 2025N3D-VLM: Native 3D Grounding Enables Accurate Spatial Reasoning in Vision-Language Modelscited by 29
  6. 2025SpatialScore: Towards Comprehensive Evaluation for Spatial Intelligencecited by 24
  7. 2025SpaceTools: Tool-Augmented Spatial Reasoning via Double Interactive RLcited by 21
  8. 2025SpaceVista: All-Scale Visual Spatial Reasoning from mm to kmcited by 19
  9. 2025SpatialThinker: Reinforcing Scene Graph-Grounded Spatial Reasoning via Dense Rewardscited by 16
  10. 2025SpinBench: Perspective and Rotation as a Lens on Spatial Reasoning in VLMscited by 14
  11. 2026pySpatial: Generating 3D Visual Programs for Zero-Shot Spatial Reasoningcited by 14
  12. 2025Struct2D: A Perception-Guided Framework for Spatial Reasoning in MLLMscited by 10
38 more
  1. 2026SceneTok: A Compressed, Diffusable Token Space for 3D Scenescited by 10
  2. 2026MLLM-4D: Towards Visual-based Spatial-Temporal Intelligencecited by 10
  3. 2025COOPER: A Unified Model for Cooperative Perception and Reasoning in Spatial Intelligencecited by 9
  4. 2025Beyond Flatlands: Unlocking Spatial Intelligence by Decoupling 3D Reasoning from Numerical Regressioncited by 8
  5. 2026PoseVLA: Universal Pose Pretraining for Generalizable Vision-Language-Action Policiescited by 7
  6. 2025Vision-Language Models as Differentiable Semantic and Spatial Rewards for Text-to-3D Generationcited by 6
  7. 2025Toward Ambulatory Vision: Learning Visually-Grounded Active View Selectioncited by 6
  8. 2026Echo-Memory: A Controlled Study of Memory in Action World Modelscited by 6
  9. 2026Allocentric Perceiver: Disentangling Allocentric Reasoning from Egocentric Visual Priors via Frame Instantiationcited by 5
  10. 2025Video2Layout: Recall and Reconstruct Metric-Grounded Cognitive Map for Spatial Reasoningcited by 4
  11. 20263D-Layout-R1: Structured Reasoning for Language-Instructed Spatial Editingcited by 4
  12. 2026World2VLM: Distilling World Model Imagination into VLMs for Dynamic Spatial Reasoningcited by 4
  13. 2026Towards Spatial Supersensing in the Wildcited by 4
  14. 2025Perceptual Taxonomy: Evaluating and Guiding Hierarchical Scene Reasoning in Vision-Language Modelscited by 3
  15. 2026MultihopSpatial: Multi-hop Compositional Spatial Reasoning Benchmark for Vision-Language Modelcited by 3
  16. 2026HiSpatial: Taming Hierarchical 3D Spatial Understanding in Vision-Language Modelscited by 3
  17. 2026OpenWorldLib: A Unified Codebase and Definition of Advanced World Modelscited by 3
  18. 2026LychSim: A Controllable and Interactive Simulation Framework for Vision Researchcited by 3
  19. 2026SPACENUM: Revisiting Spatial Numerical Understanding in VLMscited by 3
  20. 2025Pursuing Minimal Sufficiency in Spatial Reasoningcited by 2
  21. 20254D-RGPT: Toward Region-level 4D Understanding via Perceptual Distillationcited by 2
  22. 2026Smooth Operator: Smooth Verifiable Reward Activates Spatial Reasoning Ability of Vision-Language Modelcited by 2
  23. 2026XEmbodied: A Foundation Model with Enhanced Geometric and Physical Cues for Large-Scale Embodied Environmentscited by 2
  24. 2026AlloSpatial: Agentic Harness Framework for Spatial Reasoning in Foundation Modelscited by 2
  25. 2026Reason, Then Re-reason: Cross-view Revisiting Improves Spatial Reasoningcited by 2
  26. 2026HumanMoveVQA: Can Video MLLMs reason about human movement in videos?cited by 2
  27. 2026OmniView-Space: Reinforcing Spatial Reasoning via Multi-Perspective Spatial Mappingcited by 2
  28. 2025BeliefNest: A Joint Action Simulator for Embodied Agents with Theory of Mindcited by 1
  29. 2025DIV-Nav: Open-Vocabulary Spatial Relationships for Multi-Object Navigationcited by 1
  30. 20263D-RFT: Reinforcement Fine-Tuning for Video-based 3D Scene Understandingcited by 1
  31. 2026EgoReasoner: Learning Egocentric 4D Reasoning via Task-Adaptive Structured Thinkingcited by 1
  32. 2026World2Mind: Cognition Toolkit for Allocentric Spatial Reasoning in Foundation Modelscited by 1
  33. 2026From Where Things Are to What They Are For: Benchmarking Spatial-Functional Intelligence in Multimodal LLMscited by 1
  34. 2026An Enigma of Artificial Reason: Investigating the Production-Evaluation Gap in Large Reasoning Modelscited by 1
  35. 2026Reinforcing Dual-Path Reasoning in Spatial Vision Language Modelscited by 1
  36. 2026From Hallucination to Grounding: Diagnosing Visual Spatial Intelligence via CRISPcited by 1
  37. 2026From Steering to Pedalling: Do Autonomous Driving VLMs Generalize to Cyclist-Assistive Spatial Perception and Planning?cited by 0
  38. 2026SPATIALALIGN: Aligning Dynamic Spatial Relationships in Video Generationcited by 0

50 most cited of 68 All 68 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026SpatiaLab: Can Vision-Language Models Perform Spatial Reasoning in the Wild?
  2. 2025SpatialReasoner: Active Perception for Large-Scale 3D Scene Understanding
  3. 2026Do 3D Large Language Models Really Understand 3D Spatial Relationships?
  4. 2026SpatiaLQA: A Benchmark for Evaluating Spatial Logical Reasoning in Vision-Language Models
  5. 2025Mind the Gap: Benchmarking Spatial Reasoning in Vision-Language Models
  6. 2025Enhancing Spatial Reasoning through Visual and Textual Thinking
  7. 2025Spatial Understanding from Videos: Structured Prompts Meet Simulation Data
  8. 2026SpatialSpeak: QA-Native Reconstruction with Local and Global Context for Spatial Chain-of-Thought Reasoning