AI Paper Atlas

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

AgentVLN: Towards Agentic Vision-and-Language Navigation

Zihao Xin, Wentong Li, Yixuan Jiang, Ziyuan Huang, Bin Wang, Piji Li, Jianke Zhu, Jie Qin, Sheng-Jun Huang

ECCV 2026 · Vision-language and multimodal › Robot action and navigation · approach: Reasoning-enhanced vision-language navigation · 11 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 VLM-as-Brain paradigm with cross-space representation mapping (ablation)AgentVLN-3B · R2R-CE valSR: 38.60 -> 59.70Improved
add context-driven fine-grained self-correction (CDFG) (ablation)AgentVLN-3B · R2R-CE valSR: 59.70 -> 65.60Improved
add QD-PCoT (ablation)AgentVLN-3B · R2R-CE valSR: 65.60 -> 67.20Improved
add QD-PCoT (ablation)AgentVLN-3B · R2R-CE valSPL: 63.40 -> 64.70Improved
use 8 frames of temporal context instead of none (ablation)AgentVLN-3B · R2R-CE valSR: 56.7 -> 67.2Improved
extend history window beyond 8 frames (ablation)AgentVLN-3B · R2R-CE valSR / SPLHurt
AgentVLN-3B vs EfficientVLN-4B (which uses monocular depth estimation module)AgentVLN-3B · R2R-CE valSR: 64.2 -> 67.2Improved
AgentVLN-3B vs EfficientVLN-4BAgentVLN-3B · R2R-CE valOS: 73.7 -> 73.5Hurt

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