AI Paper Atlas

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

TwinVLA: Data-Efficient Bimanual Manipulation with Twin Single-Arm Vision-Language-Action Models

Hokyun Im, Euijin Jeong, Andrey Kolobov, Jianlong Fu, Youngwoon Lee

ICLR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 7 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 joint attention between the twin VLMs (ablation)TwinVLA · real-world and Tabletop-Sim Easysuccess rate: -27.0% real, -4.0% sim when removedImproved
add attention re-weighting (ablation)TwinVLA · real-world and Tabletop-Sim Easysuccess rate: -1.1% sim, -4.0% real when removed; initial loss +40%Improved
add MoE integration for shared inputs (ablation)TwinVLA · real-world and Tabletop-Sim Easysuccess rate: -1.1% sim, -5.0% real when removed; VRAM +21%Improved
add single-arm OXE pretraining (ablation)TwinVLA · real-world and Tabletop-Sim Easysuccess rate: -4.6% sim, -46.0% real when from scratchImproved
twin structure instead of monolithic RDT-1BTwinVLA · real world, simulation, language following (avg)success rate: +26.0% real, +5.0% sim, +21.8% languageImproved
use TwinVLA instead of RDT-1BRoboTwin 2.0 Hard · RoboTwin 2.0 Hardsuccess rate: -3.72% vs RDT-1BHurt
use TwinVLA instead of RDT-1BRoboTwin 2.0 Easy · RoboTwin 2.0 Easysuccess rate: +7.48%Improved
use TwinVLA instead of RDT-1BTabletop-Sim Hard · Tabletop-Sim Hardsuccess rate: +3.3%Improved
TwinVLA vs pi0 under distractorsFold towel (real) · Fold towel with distractorssuccess rate (%): 25.0 vs pi0 60.0Hurt
pretrain SingleVLA on OXE before LIBERO fine-tuning (ablation)SingleVLA · LIBERO avgsuccess rate (%): 74.0 -> 86.0Improved

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