AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

CabinSI: Omni-Cabin Spatial Reasoning through Explicit Visual Cognitive Maps

Mengxue Qu, Hengrui Hu, Mingming Ma, Ming Lei, Jie Gao, Henghui Ding, Yao Zhao, Kenn wu, Yunchao Wei

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Spatial reasoning in VLMs · 0 citations (OpenAlex)

Read the paper

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 cognitive map frameworkQwen2.5-VL-7B · RelCabin overallaccuracy (%): 25.31 -> 41.17Improved
add cognitive map frameworkQwen3-VL-4B · RelCabin overallaccuracy (%): 27.17 -> 44.46Improved
add cognitive map frameworkQwen3-VL-8B · RelCabin overallaccuracy (%): 33.59 -> 45.92Improved
add cognitive map frameworkQwen3-VL-8B · RelCabin in-cabin multi-viewaccuracy (%): 19.80 -> 24.75Improved
add cognitive map frameworkQwen2.5-VL-7B · RelCabin in-cabin multi-viewaccuracy (%): 16.83 -> 26.73Improved
filter to target view only (ablation)Rule-extracted image CogMap · RelCabin out-cabin multi-viewaccuracy (%): 32.99 -> 33.15No clear effect
replace rule-based object extraction with GPT-extracted objects (ablation)CogMap image only · RelCabin out-cabin multi-viewaccuracy (%): 37.48 -> 47.46Improved
remove raw multi-view images, give cognitive map image only (ablation)GPT-extracted CogMap · RelCabin out-cabin multi-viewaccuracy (%): 35.06 -> 47.46Improved
add text form of map alongside image (ablation)Target-view GPT-extracted CogMap · RelCabin out-cabin multi-viewaccuracy (%): 35.06 -> 33.20Hurt
add proposal generation with selection (ablation)Qwen2.5-VL-7B · RefCabin in-cabin multi-viewlocalization accuracy (%): 14.95 -> 22.16Improved
add cognitive map construction on top of proposals (ablation)Qwen2.5-VL-7B · RefCabin in-cabin multi-viewlocalization accuracy (%): 22.16 -> 24.23Improved
add cognitive map construction on top of proposals (ablation)Qwen2.5-VL-7B · RefCabin in-cabin single-viewlocalization accuracy (%): 48.26 -> 52.90Improved

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