AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

VIEW2SPACE: Studying Multi-View Visual Reasoning from Sparse Observations

Fucai Ke, Zhixi Cai, Boying Li, Long Chen, Beibei Lin, Weiqing Wang, Pari Delir Haghighi, Gholamreza Haffari, Hamid Rezatofighi

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Spatial reasoning in VLMs · 9 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
add grounded CoT with visual evidence (vs instruction tuning) (ablation)Qwen3-VL-4B · VIEW2SPACE-v1MCQ accuracy (%): 36.50 -> 64.93Improved
add grounded CoT with visual evidence (vs instruction tuning) (ablation)Qwen3-VL-4B · VIEW2SPACE-v1grounding mIoU (%): 47.58 -> 69.34Improved
add bounding-box grounding to CoT (vs standard CoT) (ablation)Qwen3-VL-4B · VIEW2SPACE-v1grounding mIoU (%): 49.77 -> 69.34Improved
add standard CoT training (vs instruction tuning) (ablation)Qwen3-VL-4B · VIEW2SPACE-v1MCQ accuracy (%): 36.50 -> 59.57Improved
add standard CoT training (vs instruction tuning) (ablation)Qwen3-VL-4B · VIEW2SPACE-v1grounding mIoU (%): 47.58 -> 49.77Improved
train with CoT but answer directly at test (ablation)Qwen3-VL-4B grounded CoT · VIEW2SPACE-v1MCQ accuracy (%): 64.93 -> 33.95Hurt
train on VIEW2SPACE and test on MindCube without tuningQwen3-VL-4B grounded CoT · MindCube-Tinyaccuracy (%): 60.76 -> 70.00Improved
enforce explicit thinking at inference (ablation)open-source MLLMs · VIEW2SPACE-v1MCQ accuracy (%): Qwen3-VL-8B 31.79 -> 37.41; Qwen3-VL-4B 32.79 -> 35.19; Gemma 32.50 -> 31.14Mixed
upgrade GPT-5 to GPT-5.2GPT · VIEW2SPACE-v1MCQ accuracy (%): 59.86 -> 37.07Hurt
train on 300K data instead of smaller data / scale dataset beyond 50K (ablation)Qwen3-VL-2B/4B · VIEW2SPACE reasoning difficulty analysisaccuracyNo clear effect
use open-source spatial models vs general MLLMsSpatial models · VIEW2SPACE-v1MCQ accuracy (%): Spatial-MLLM 27.93 vs Qwen3-VL-8B 37.41No 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