AI Paper Atlas

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

Harnessing Input-Adaptive Inference for Efficient VLN

Dongwoo Kang, Akhil Perincherry, Zachary Coalson, Aiden Gabriel, Stefan Lee, Sanghyun Hong

ICCV 2025 · Vision-language and multimodal › Robot action and navigation · approach: Reasoning-enhanced vision-language navigation · 2 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
apply existing early-exit method MuE to the HAMT ViTHAMT · R2R val unseenSR: 66.16 -> 43.93Hurt
apply MuEHAMT · R2R val unseenGFLOPs: 4763.24 -> 4409.62No clear effect
add k-extension to process only navigable views plus k neighbors (ablation)HAMT · REVERIE val unseenGFLOPs (+C): 5434.71 -> 3121.20 (SR 31.81 -> 24.96)Improved
add LSH caching on top of k-extension (ablation)HAMT · REVERIE val unseenGFLOPs (+C): 3121.20 -> 2359.72Improved
add adaptive thresholds on top of k-extension (ablation)HAMT · REVERIE val unseenGFLOPs (+C): 3121.20 -> 2723.01 (SR 24.96 -> 24.60)Improved
increase k from 1 to 6 (ablation)k-extension · R2R val unseenSR: 54.32 -> 62.84 (GFLOPs 1250.65 -> 2414.46)Improved
increase early-exit aggressiveness A from 0 to 0.022 (ablation)adaptive thresholding · R2R val unseenSR: 61.17 -> 55.60 (GFLOPs 2013.48 -> 1768.85)Hurt
swap cosine similarity for SSIM, FSIM, LPIPS or ViT activation in LSH (ablation)LSH · R2R val unseenSR: 60.41 -> 60.49 (LPIPS)No clear effect
add LSH caching to scan-only SGM in continuous VLN (ablation)VLN-CE BERT · R2R-CE val unseenSR: 39.86 -> 37.14 (GFLOPs 2396.54 -> 1741.33)Hurt
apply scan-only subgoal generation module to predict navigable viewsVLN-CE BERT · R2R-CE val unseenGFLOPs: 18074.05 -> 2396.54Improved
apply visual corruption (Speckle Noise) to efficient agentefficient HAMT · R2R val unseenSR: 60.41 -> 40.87Hurt

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