AI Paper Atlas

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

LEAP-VLA: Latent-Enhanced Action Prototyping via Continuous Residual Latent Spaces for Vision-Language-Action Models

Bo-Yun Yu, Kuan-Chuan Peng, Jun Hsieh

ECCV 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
predict latent zq targets instead of raw actions (ablation)LEAP-VLA aligner with linear head · LIBERO-Spatialsuccess rate (%): 92.4 -> 94.6 (+2.2%)Improved
replace linear head with Prototype-Aware Output Head (ablation)LEAP-VLA aligner · LIBERO-Spatialsuccess rate (%): 94.6 -> 98.4 (+3.8%)Improved
use multi-level residual quantization instead of single-level soft assignment (ablation)MSRQ · LIBERO-SpatialMAE (x10^-3): 2.5 -> 1.5Improved
use L2-based soft assignment instead of discrete hard assignment (MSRQ-single vs discrete multi-level) (ablation)MSRQ-single · LIBERO-SpatialMAE (x10^-3): 2.5 vs 7.7 (SimRVQ), 9.0 (RVQ-Commit)Improved
use soft residual levels instead of discrete residual levels for utilization (ablation)MSRQ vs RVQ-Commit · LIBERO-Spatial level-0 utilizationentropy ratio (%): 51.9% -> 99.5%Improved
use bidirectional Transformer with CNN (BiCT) instead of causal Transformer (ablation)MSRQ encoder/decoder · LIBERO-SpatialMAE (x10^-3): 71.9 -> 1.5Improved
use BiCT instead of CNN only (ablation)MSRQ encoder/decoder · LIBERO-SpatialMAE (x10^-3): 8.5 -> 1.5Improved
use BiCT instead of MLP only (ablation)MSRQ encoder/decoder · LIBERO-SpatialMAE (x10^-3): 15.4 -> 1.5Improved
use 1B LEAP-VLA instead of larger robot-pretrained VLAsLEAP-VLA · LIBERO averagesuccess rate (%): 92.7 (UD-VLA 8.5B) -> 97.1Improved
use LEAP-VLA without robot pretraining vs OpenVLA-OFT with robot pretrainingLEAP-VLA · CALVIN ABC->Daverage sequence length: 4.10 -> 4.28Improved
use LEAP-VLA instead of VLM4VLA (VLM-only)LEAP-VLA · CALVIN ABC->Daverage sequence length: 4.06 -> 4.28Improved
use LEAP-VLA instead of pi0 on LIBERO-LongLEAP-VLA · LIBERO-Longsuccess rate (%): 85.2 -> 93.0Improved

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