AI Paper Atlas

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

Language and Visual Entity Relationship Graph for Agent Navigation

Yicong Hong, Cristian Rodriguez, Yuankai Qi, Qi Wu, Stephen Gould

NeurIPS 2020 · Vision-language and multimodal › Robot action and navigation · approach: Reasoning-enhanced vision-language navigation · 20 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
Model scene, object, and direction clues with language and visual relationship graphsVLN agent · R2R test unseenSPL: 0.47 -> 0.52Improved
Model scene, object, and direction clues with language and visual relationship graphsVLN agent · R4R validation unseenSDTW: 0.13 -> 0.34Improved
Add specialized scene, object, and direction clues (ablation)VLN baseline · R2R validation unseenSPL: 0.48 -> 0.51Improved
Add object clues to a graph model with scene and direction clues (ablation)Model 2 · R2R validation unseenSPL: 0.50 -> 0.53Improved
Add scene clues to a graph model with object and direction clues (ablation)Model 3 · R2R validation unseenSPL: 0.46 -> 0.53Improved
Add graph relationship modeling to scene, object, and direction clues (ablation)Model 4 · R2R validation unseenSPL: 0.51 -> 0.53Improved
Add language graph edges when visual graph edges are present (ablation)Model 5 · R2R validation unseenSPL: 0.50 -> 0.53Improved
Add object visual clues (ablation)Model 2 · R2R validation unseen, instructions with 1 objectsuccess rate: 0.504 -> 0.504No clear effect
Add object visual clues (ablation)Model 2 · R2R validation unseen, instructions with 9 objectssuccess rate: 0.405 -> 0.459Improved

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