AI Paper Atlas

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

LARA: Latent Action Representation Alignment for Vision-Language-Action Models

Mengya Liu, Baoxiong Jia, Jiangyong Huang, Jingze Zhang, Siyuan Huang

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 LARA latent action representation alignment to a vanilla DiT VLA (ablation)LARA (DiT-only) · LIBERO (OXE-constrained) averagesuccess rate (%): 84.4 -> 88.6Improved
add LARA alignment to a vanilla DiT VLA (ablation)LARA (DiT-only) · SIMPLER-ENV (OXE-constrained) averagesuccess rate (%): 55.8 -> 65.2Improved
add LARA alignment to a vanilla DiT VLA (ablation)LARA (DiT-only) · LIBERO-Long (OXE-constrained)success rate (%): 76.5 -> 86.0Improved
add LARA alignment to a vanilla DiT VLA (ablation)LARA (DiT-only) · SIMPLER-ENV Drawer (OXE-constrained)success rate (%): 21.0 -> 29.5Improved
add LARA alignment to a vanilla DiT VLA (ablation)LARA (DiT-only) · SIMPLER-ENV Move (OXE-constrained)success rate (%): 84.0 -> 83.7No clear effect
post-train GR00T-N1.6 with LARA using an OXE-pretrained LAMGR00T-N1.6 · LIBERO (unconstrained) averagesuccess rate (%): 95.0 -> 95.6Improved
post-train GR00T-N1.6 with LARAGR00T-N1.6 · SIMPLER-ENV (unconstrained) averagesuccess rate (%): 78.9 -> 79.9Improved
replace vanilla LAM with LARA-refined LAM as pseudo-labelsMoto-GPT · SIMPLER-ENV (Moto-GPT, OXE Fractal) averagesuccess rate (%): 42.8 -> 49.5Improved
apply alignment at layer L instead of L-2 (ablation)pi0.5-LARA · LIBERO average (pi0.5 post-training)success rate (%): 91.2 -> 97.8Improved
apply LARA at final layer to pi0.5pi0.5 · LIBERO average (pi0.5 post-training)success rate (%): 96.9 -> 97.8Improved
vary alignment depth over layers 4, 8, L-2, L (ablation)GR00T-N1.6-LARA · LIBERO-Long (unconstrained)success rate (%)Mixed
jointly optimize LAM with the VLA instead of freezing the LAM (ablation)GR00T-N1.6-LARA · LIBERO-Long (unconstrained)success rate (%)Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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