AI Paper Atlas

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

RA-VLA: Retrieval-Augmented VLA for Test-Time Adaptation

Sanghwan Jang, Minjin Jeon, Minsoo Kim, Seong Jin Choi, Dongha Kim, Hwanjo Yu

ICML 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 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 contextual adherence loss (ablation)RA-VLA · LIBERO-Goalsuccess rate: 0.098 -> 0.532Improved
add contextual adherence loss (ablation)RA-VLA · LIBERO-GoalS_ctx: 0.0353 -> 0.3639Improved
replace off-the-shelf retriever with action-aware retriever (ablation)RA-VLA · LIBERO-Goalsuccess rate: 10.2% -> 53.2%Improved
use action-aware retriever instead of SigLIP 2 (ablation)RAEA · LIBERO averagesuccess rate: 0.0450 -> 0.1235Improved
use action-aware retriever instead of SigLIP 2 (ablation)RICL_G · LIBERO averagesuccess rate: 0.0885 -> 0.1295Improved
use action-aware retriever instead of SigLIP 2 (ablation)RICL_R · LIBERO averagesuccess rate: 0.0465 -> 0.2085Improved
RA-VLA instead of best baseline RICL_RLIBERO unseen suite · LIBERO averagesuccess rate: 20.9% -> 38.5%Improved
RA-VLA instead of best baseline RICL_RUR5e real robot · UR5e averagesuccess rate: 35.4% -> 56.3%Improved
increase buffer size N from 1 to 4 (ablation)RA-VLA · LIBERO-Goalsuccess rate: 0.482 -> 0.552Improved
increase retrieved segments K from 1 to 4 (ablation)RA-VLA · LIBERO-Goalsuccess rate: 0.532 -> 0.546No clear effect
decoupled per-segment encoding vs concatenating segments in promptRA-VLA · inference latency vs KlatencyImproved

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