AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

NavCoT: Boosting LLM-Based Vision-and-Language Navigation via Learning Disentangled Reasoning

Bingqian Lin, Yunshuang Nie, Ziming Wei, Jiaqi Chen, Shikui Ma, Jianhua Han, Hang Xu, Xiaojun Chang, Xiaodan Liang

TPAMI 2025 · Vision-language and multimodal › Robot action and navigation · approach: Pretraining and data augmentation for VLN · 118 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 the Future Imagination task to action prediction (pretrain+finetune) (ablation)NavCoT LLaMA 2 · R2R Val Unseen subsetSR: 21.11 -> 24.44Improved
add the Visual Information Filter task to action prediction (pretrain+finetune) (ablation)NavCoT LLaMA 2 · R2R Val Unseen subsetSPL: 14.82 -> 19.08Improved
add VIF on top of AP+FI (ablation)NavCoT LLaMA 2 · R2R Val Unseen subsetSR: 24.44 -> 30.00Improved
add FI on top of AP+VIF (ablation)NavCoT LLaMA 2 · R2R Val Unseen subsetSR: 23.33 -> 30.00Improved
use NavCoT prompt instead of original CoT prompt, zero-shot LLaMA 2 (ablation)LLaMA 2 · R2R Val Unseen subsetSR: 3.33 -> 16.67Improved
add in-domain training to NavCoT prompt (ablation)NavCoT LLaMA 2 · R2R Val Unseen subsetSR: 16.67 -> 53.33Improved
use LLaMA 2 instead of LLaMA-Adapter backbone (ablation)NavCoT · R2R Val Unseen subsetSR: 28.89 -> 53.33Improved
use LLaMA 2 instead of GPT-4 in zero-shot NavCoT (ablation)NavCoT zero-shot · R2R Val Unseen subsetSR: 15.56 -> 16.67Improved
let the LLM choose the action instead of CLIP matching the imagination (ablation)NavCoT LLaMA 2 · R2R Val Unseen subsetSR: 32.23 -> 53.33Improved
add 12000 R2R augmentation samplesNavCoT LLaMA 2 · R2R Val UnseenSR: 36.40 -> 40.23Improved
replace direct action prediction with NavCoT reasoningLLaMA 2 7B · RxR English Val UnseenSR: 20.46 -> 24.52Improved
replace direct action prediction with NavCoT reasoningLLaMA 2 7B · REVERIE Val UnseenSR: 3.12 -> 9.20Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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