AI Paper Atlas

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

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Spatial reasoning in VLMs · 128 citations (OpenAlex)

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 changedWhereMeasuredEffect
fine-tune on 43K multi-perspective spatial QA dataQwen2.5-VL-3B · ViewSpatial-Benchoverall accuracy (%): 35.85 -> 82.09Improved
fine-tune on multi-perspective dataQwen2.5-VL-7B · ViewSpatial-Benchoverall accuracy (%): 36.85 -> 83.01Improved
fine-tune on multi-perspective dataInternVL2.5-2B · ViewSpatial-Benchoverall accuracy (%): 34.98 -> 76.45Improved
train with direct-answer instead of multiple-choice format (ablation)MVSM Qwen2.5-VL-3B · ViewSpatial-Benchoverall accuracy (%): 82.09 -> 79.34No clear effect
apply MVSM training for transferQwen2.5-VL-3B · VSI-Bench Route Planaccuracy (%): 21.90 -> 31.44Improved
apply MVSM training for transferQwen2.5-VL-3B · VSI-Bench Rel Diraccuracy (%): 46.00 -> 46.93No clear effect
apply MVSM training for transfer to indoor scenesQwen2.5-VL-3B · VSI-App indooraccuracy (%): 18.00 -> 41.00Mixed
apply MVSM training for transfer to outdoor scenesQwen2.5-VL-3B · VSI-App outdooraccuracy (%): 27.00 -> 36.00Mixed
proprietary models vs random chance on perspective tasksGPT-4o · ViewSpatial-Benchoverall accuracy (%): 34.98 vs 26.33 randomNo clear effect

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy