AI Paper Atlas

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

AMCoNav: Asynchronous Multi-module Collaborative Framework for Embodied Visual Navigation

Jiaquan Yan, Fang Zhao, Yushi Chen, Long wang, Haiyong Luo, Dan Luo

ECCV 2026 · Vision-language and multimodal › Robot action and navigation · approach: Pretraining and data augmentation for VLN · 0 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 global guidance (MSC + ZLMM) to MTU3D (ablation)MTU3D · HM3D-OVON Val UnseenSR (%): 40.8 -> 44.2Improved
add global guidance (MSC + ZLMM) to MTU3D (ablation)MTU3D · HM3D-OVON Val Seen SynonymsSR (%): 45.0 -> 53.3Improved
add BPDFM on top of global guidance (ablation)AMCoNav · HM3D-OVON Val UnseenSR (%): 44.2 -> 53.3Improved
add BPDFM on top of global guidance (ablation)AMCoNav · HM3D-OVON Val Seen SynonymsSPL (%): 17.5 -> 20.9Improved
add BPDFM on top of global guidance (ablation)AMCoNav · HM3D-OVON Val SeenSPL (%): 15.8 -> 15.6No clear effect
use AMCoNav vs MTU3D (dual verification)AMCoNav · HM3D-OVON Val SeenSPL (%): 23.6 -> 15.6Hurt
use AMCoNav vs MTU3DAMCoNav · SG3D-NavSPL (%): 16.5 -> 11.3Hurt
use AMCoNav vs MTU3DAMCoNav · GOAT-Bench Val UnseenSR (%): 47.2 -> 47.2No clear effect
run large-model TDM asynchronously instead of synchronously (ablation)AMCoNav · Runtime per episode (100 episodes)seconds: 174.1 -> 142.5Improved
set verification step threshold c1 to 500 instead of 300 (ablation)AMCoNav · HM3D-OVON Val Seen (50 episodes)SR (%): 60.0 -> 52.0Hurt
set verification step threshold c1 to 200 instead of 300 (ablation)AMCoNav · HM3D-OVON Val Seen (50 episodes)SR (%): 60.0 -> 56.0Hurt
replace GPT-5 with GPT-4o as ZLMM (ablation)AMCoNav · HM3D-OVON Val Seen (50 episodes)SR (%): 60.0 -> 56.0Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 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