AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Visual Spatial Tuning

Rui Yang, ziyu zhu, Yanwei Li, Jingjia Huang, Shen Yan, Siyuan Zhou, Zhe Liu, Xiangtai Li, Shuangye Li, Wenqian Wang, Yi Lin, Hengshuang ZHAO

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Spatial reasoning in VLMs · 103 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 single-image scene caption data (ablation)VST-3B SFT · spatial benchmarksoverall improvement: 2.3%Improved
add depth and distance data (ablation)VST-3B SFT · CVBench-3Daccuracy: 10.4%Improved
add multi-view correspondence data (ablation)VST-3B SFT · BLINKaccuracy: 1.8%Improved
add multi-image 3D object detection data (ablation)VST-3B SFT · MMSI-Benchaccuracy: 2.1%Improved
add general video data (ablation)VST-3B SFT · VSI-Benchaccuracy: 38.4 -> 38.1Hurt
add VST-video data (ablation)VST-3B SFT · VSI-Benchaccuracy: 38.4 -> 54.7Improved
generate CoT conditioned on BEV image (ablation)Qwen2.5-VL-7B cold start · MMSI-Benchoverall accuracy: 30.0 -> 31.1Improved
add CoT inference after cold start (ablation)Qwen2.5-VL-7B cold start · MMSI-Benchoverall accuracy: 33.6 -> 31.7Hurt
add RL stage after cold start (ablation)VST cold-start model · MMSI-Benchoverall accuracy: 31.7 -> 35.3Improved
add FoV unification (ablation)VST 3D detection · ARKitScenesAP@15: 40.9 -> 42.7Improved
use multi-turn instead of single-turn detection data (ablation)VST 3D detection · SUN RGB-DAP@15: 26.0 -> 28.5Improved
use IoU plus F1 reward instead of IoU plus recall (ablation)VST RL 3D detection · SUN RGB-DAP: 13.8 -> 24.4Improved

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