vulab-AI/Awesome-Spatial-VLMs

• highreadme#1

  Reposition the README opening to emphasize its role as a definitive resource

  原因：

  当前

  >A curated hub for Spatial Intelligence in Vision-Language Models. Actively maintained—watch for updates, benchmark your VLM with our evaluation code, and consider starring 🌟 and sharing if helpful. This repository is the official, community-maintained resource for our survey paper: **Spatial Intelligence in Vision-Language Models: A Comprehensive Survey**.

  复制粘贴的修复

  复制

  This is the **definitive, community-maintained hub** for Spatial Intelligence in Vision-Language Models, serving as the official resource for our comprehensive survey paper. It centralizes papers, benchmarks, and evaluation code, making it the go-to collection for researchers and developers.

  待办完成忽略
• hightopics#2

  Add more specific topics related to benchmarks, evaluation, and paper collection

  原因：

  当前

  awesome-list, awesome-spatial-vlms, mllm, spatial-intelligence, spatial-reasoning, survey, vision-language-model, vlm

  复制粘贴的修复

  复制

  awesome-list, awesome-spatial-vlms, mllm, spatial-intelligence, spatial-reasoning, survey, vision-language-model, vlm, vlm-benchmarks, vlm-evaluation, paper-collection, research-resources

  待办完成忽略
• mediumreadme#3

  Add a dedicated section in the README for Benchmarks & Evaluation

  原因：

  复制粘贴的修复

  复制

  ## Benchmarks & Evaluation
  
  This repository provides comprehensive resources for benchmarking and evaluating Spatial Vision-Language Models.
  *   **Evaluation Code:** Explore our official evaluation framework [here](https://github.com/vulab-AI/Awesome-Spatial-VLMs/blob/main/evaluation/README.md).
  *   **Spatial VQA Datasets:** Access curated datasets for Spatial VQA [here](https://github.com/vulab-AI/Awesome-Spatial-VLMs/blob/main/data_benchmark/Dataset_SVQA.md).
  *   **Spatial VQA Benchmarks:** Review detailed benchmarks for Spatial VQA [here](https://github.com/vulab-AI/Awesome-Spatial-VLMs/blob/main/data_benchmark/Benchmark_SVQA.md).

  待办完成忽略

本次扫描解析到的品类 GEO 通道：google/gemini-2.5-flash, deepseek/deepseek-v4-flash

• 品类问题

  How can I find resources and papers on spatial reasoning in vision-language models?

  你：未被推荐
  AI 推荐顺序：

  1. Google Scholar
  2. arXiv.org
  3. Semantic Scholar
  4. Papers With Code
  5. CVPR
  6. ICCV
  7. ECCV
  8. NeurIPS
  9. ICLR
  10. ACL
  11. GitHub
  12. Stanford Vision and Learning Lab
  13. UC Berkeley AI Research (BAIR)
  14. Allen Institute for AI (AI2)
  15. Meta AI Research (FAIR)
  16. Google DeepMind
  17. Google AI

  AI 推荐了 17 个替代方案，却始终没点名 vulab-AI/Awesome-Spatial-VLMs。这就是要补上的差距。

  查看 AI 完整回答
• 品类问题

  Where can I find benchmarks and evaluation methods for spatial intelligence in VLMs?

  你：未被推荐
  AI 推荐顺序：

  1. GQA
  2. CLEVR
  3. NLVR2
  4. OK-VQA
  5. Touchdown
  6. ALFRED
  7. VISPROG

  AI 推荐了 7 个替代方案，却始终没点名 vulab-AI/Awesome-Spatial-VLMs。这就是要补上的差距。

  查看 AI 完整回答

• Metadata completeness

  pass

• README presence

  pass

• Compared to common alternatives in this category, what is the core differentiator of vulab-AI/Awesome-Spatial-VLMs?

  pass
  ⚠AI 未点名 vulab-AI/Awesome-Spatial-VLMs —— 很可能在说另一个项目

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？

• If a team adopts vulab-AI/Awesome-Spatial-VLMs in production, what risks or prerequisites should they evaluate first?

  pass
  ✓AI 明确点名了 vulab-AI/Awesome-Spatial-VLMs

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？

• In one sentence, what problem does the repo vulab-AI/Awesome-Spatial-VLMs solve, and who is the primary audience?

  pass
  ⚠AI 未点名 vulab-AI/Awesome-Spatial-VLMs —— 很可能在说另一个项目

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？