AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

CoNav: Collaborative Cross-Modal Reasoning for Embodied Navigation

Haihong Hao, Mingfei Han, Changlin Li, Zhihui Li, Xiaojun Chang

TPAMI 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Spatial reasoning in VLMs · 7 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 Cross-Modal Belief Alignment fine-tuning on top of sharing 3D-text hypotheses (ablation)CoNav · R2R Val-USPL: 56 -> 62Improved
add Cross-Modal Belief Alignment fine-tuning (ablation)CoNav · ScanQA Val-UROUGE: 35.7 -> 38.9Improved
share 3D-text hypotheses without alignment to image-text agent (ablation)image-text agent · R2R Val-USPL: 57 -> 56Hurt
share 3D-text hypotheses without alignment (ablation)image-text agent · ScanQA Val-UROUGE: 36.7 -> 35.7Hurt
add 3D-text model guidance with alignment to image-text agent (ablation)image-text agent · R2R Val-USR: 64 -> 69Improved
replace post-process CEO alignment (Qwen2.5-7B) by trained alignment (ablation)CoNav · R2R Val-USPL: 51 -> 62Improved
train alignment instead of post-process CEO alignment (ablation)CoNav · ScanQA Val-UROUGE: 35.2 -> 38.9Improved
use model-level fusion instead of feature-level fusion (ablation)CoNav · ScanQA Val-UROUGE: 31.7 -> 38.9Improved
use model-level fusion instead of feature-level fusion (ablation)CoNav · ScanQA Val-UMETEOR: 13.9 -> 15.2Improved
add curriculum stage 2 (3D-FRONT) after stage 1 for 3D-text model (ablation)3D-text model · ScanQA Val-UMETEOR: 8.39 -> 10.1Improved
add curriculum stage 3 (real scans) after stages 1-2 (ablation)3D-text model · ScanQA Val-UMETEOR: 10.1 -> 14.5Improved
use point-cloud text model alone vs image-text model alone for spatial reasoning (ablation)single-modality models · ScanQA Val-UMETEOR: 14.3 -> 14.5Improved

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