mlsanityData to 7 Oct 2026

SpaRE: Enhancing Spatial Reasoning in Vision-Language Models with Synthetic Data

arXiv 2504.20648 ↗First posted 29 Apr 2025cs.CV cs.AIcited by 19

On the map: LLM reasoning → Spatial and embodied reasoning

also in 2 more
Abstract (arXiv)

Vision-language models (VLMs) work well in tasks ranging from image captioning to visual question answering (VQA), yet they struggle with spatial reasoning, a key skill for understanding our physical world that humans excel at. We find that spatial relations are generally rare in widely used VL datasets, with only a few being well represented, while most form a long tail of underrepresented relations. This gap leaves VLMs ill-equipped to handle diverse spatial relationships. To bridge it, we construct a synthetic VQA dataset focused on spatial reasoning generated from hyper-detailed image descriptions in Localized Narratives, DOCCI, and PixMo-Cap. Our dataset consists of 455k samples containing 3.4 million QA pairs. Trained on this dataset, our Spatial-Reasoning Enhanced (SpaRE) VLMs show strong improvements on spatial reasoning benchmarks, achieving up to a 49% performance gain on the What's Up benchmark, while maintaining strong results on general tasks. Our work narrows the gap between human and VLM spatial reasoning and makes VLMs more capable in real-world tasks such as robotics and navigation.

Key facts. Cited by 19 papers in the map (data through 7 Oct 2026). Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning. Builds on 14 of its 47 references that are in the map. First posted on arXiv 29 Apr 2025.

Builds on

14 of 47 references.

Closest ancestors · 2

Same direction, newest first

  1. 20243DSRBench: A Comprehensive 3D Spatial Reasoning Benchmark10 Dec
  2. 2023STUPD: A Synthetic Dataset for Spatial and Temporal Relation Reasoning13 Sep

Other references · 12

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  3. 2024Qwen2.5 Technical Reportcited by 4,713
  4. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,151
  5. 2021CLIPScore: A Reference-free Evaluation Metric for Image Captioningcited by 2,388
  6. 2024Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phonecited by 1,883
  7. 2023A Survey on Multimodal Large Language Modelscited by 898
  8. 2023Textbooks Are All You Needcited by 561
  9. 2023Textbooks Are All You Need II: phi-1.5 technical reportcited by 559
  10. 2024Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Modelscited by 321
  11. 2022Benchmarking Spatial Relationships in Text-to-Image Generationcited by 96
  12. 2024Do Vision-Language Models Represent Space and How? Evaluating Spatial Frame of Reference Under Ambiguitiescited by 58

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 9; 2026: 10.

19 papers.

Sort:

By area:

  1. 2025How Far are VLMs from Visual Spatial Intelligence? A Benchmark-Driven Perspectivecited by 41
  2. 2025Multimodal Spatial Reasoning in the Large Model Era: A Survey and Benchmarkscited by 31
  3. 2025Spatial Reasoning in Multimodal Large Language Models: A Survey of Tasks, Benchmarks and Methodscited by 22
  4. 2025SpatialThinker: Reinforcing Scene Graph-Grounded Spatial Reasoning via Dense Rewardscited by 14
  5. 2026From Seeing to Thinking: Decoupling Perception and Reasoning Improves Post-Training of Vision-Language Modelscited by 5
  6. 2025RAU: Reference-based Anatomical Understanding with Vision Language Modelscited by 3
  7. 2025GRAID: Enhancing Spatial Reasoning of VLMs Through High-Fidelity Data Generationcited by 3
  8. 2026World2VLM: Distilling World Model Imagination into VLMs for Dynamic Spatial Reasoningcited by 3
  9. 2025Reasoning Path and Latent State Analysis for Multi-view Visual Spatial Reasoning: A Cognitive Science Perspectivecited by 2
  10. 2026MosaicThinker: On-Device Visual Spatial Reasoning for Embodied AI via Iterative Construction of Space Representationcited by 2
  11. 2026Reason, Then Re-reason: Cross-view Revisiting Improves Spatial Reasoningcited by 2
  12. 2026PRISM: A Benchmark for Programmatic Spatial-Temporal Reasoningcited by 1
7 more
  1. 2025SpatialTraceGen: High-Fidelity Traces for Efficient VLM Spatial Reasoning Distillationcited by 0
  2. 2025The N-Body Problem: Parallel Execution from Single-Person Egocentric Videocited by 0
  3. 2026From Steering to Pedalling: Do Autonomous Driving VLMs Generalize to Cyclist-Assistive Spatial Perception and Planning?cited by 0
  4. 2026Learning to Draw ASCII Improves Spatial Reasoning in Language Modelscited by 0
  5. 2026ReRef-3D: A Benchmark for Spatial Referring Expression-Guided 3D Scene Rearrangementcited by 0
  6. 2026MVVBench: Benchmarking 4D Reasoning in Vision-Language Modelscited by 0
  7. 2026GaugeVLM: Structuring Spatial Supervision with Measured Geometric Interventionscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities
  2. 2025Mind the Gap: Benchmarking Spatial Reasoning in Vision-Language Models
  3. 2024Reasoning Paths with Reference Objects Elicit Quantitative Spatial Reasoning in Large Vision-Language Models
  4. 2026SpatiaLQA: A Benchmark for Evaluating Spatial Logical Reasoning in Vision-Language Models
  5. 2024Is A Picture Worth A Thousand Words? Delving Into Spatial Reasoning for Vision Language Models
  6. 2026SpatiaLab: Can Vision-Language Models Perform Spatial Reasoning in the Wild?
  7. 2023What's "up" with vision-language models? Investigating their struggle with spatial reasoning
  8. 2026SpatialChain: A Benchmark for Auditing Spatial Reasoning Faithfulness in VLMs