Research map / Vision-language and multimodal / Multimodal LLMs
ViewSpatial-Bench: Evaluating Multi-perspective Spatial Localization in Vision-Language Models
Dingming Li, Hongxing Li, Zixuan Wang, Yuchen Yan, Hang Zhang, Siqi Chen, Guiyang Hou, Shengpei Jiang, Wenqi Zhang, Yongliang Shen, Weiming Lu, Yueting Zhuang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| fine-tune on 43K multi-perspective spatial QA data | Qwen2.5-VL-3B · ViewSpatial-Bench | overall accuracy (%): 35.85 -> 82.09 | Improved |
| fine-tune on multi-perspective data | Qwen2.5-VL-7B · ViewSpatial-Bench | overall accuracy (%): 36.85 -> 83.01 | Improved |
| fine-tune on multi-perspective data | InternVL2.5-2B · ViewSpatial-Bench | overall accuracy (%): 34.98 -> 76.45 | Improved |
| train with direct-answer instead of multiple-choice format (ablation) | MVSM Qwen2.5-VL-3B · ViewSpatial-Bench | overall accuracy (%): 82.09 -> 79.34 | No clear effect |
| apply MVSM training for transfer | Qwen2.5-VL-3B · VSI-Bench Route Plan | accuracy (%): 21.90 -> 31.44 | Improved |
| apply MVSM training for transfer | Qwen2.5-VL-3B · VSI-Bench Rel Dir | accuracy (%): 46.00 -> 46.93 | No clear effect |
| apply MVSM training for transfer to indoor scenes | Qwen2.5-VL-3B · VSI-App indoor | accuracy (%): 18.00 -> 41.00 | Mixed |
| apply MVSM training for transfer to outdoor scenes | Qwen2.5-VL-3B · VSI-App outdoor | accuracy (%): 27.00 -> 36.00 | Mixed |
| proprietary models vs random chance on perspective tasks | GPT-4o · ViewSpatial-Bench | overall accuracy (%): 34.98 vs 26.33 random | No clear effect |
Closest papers by meaning
- InternSpatial: A Comprehensive Dataset for Spatial Reasoning in Vision-Language Models (ICLR 2026 · similarity 0.94)
- SpatialRGPT: Grounded Spatial Reasoning in Vision-Language Models (NeurIPS 2024 · similarity 0.93)
- SpinBench: Perspective and Rotation as a Lens on Spatial Reasoning in VLMs (ICLR 2026 · similarity 0.93)
- Seeing Across Views: Benchmarking Spatial Reasoning of Vision-Language Models in Robotic Scenes (ICLR 2026 · similarity 0.92)
- Spatial Reasoning with Vision-Language Models in Ego-Centric Multi-View Scenes (ICLR 2026 · similarity 0.92)
- Is A Picture Worth A Thousand Words? Delving Into Spatial Reasoning for Vision Language Models (NeurIPS 2024 · similarity 0.92)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
