AI Paper Atlas

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

Hybrid Training for Vision-Language-Action Models

Pietro Mazzaglia, Cansu Sancaktar, Markus Peschl, Daniel Dijkman

ICLR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 4 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
train with Hybrid Training (HyT) instead of standard OpenVLA fine-tuningOpenVLA 7B · Real-world tasks overallsuccess rate: 41% -> 63%Improved
train with HyT instead of OpenVLAOpenVLA 7B · Real-world out-of-distribution taskssuccess rate: 29% -> 54%Improved
train with HyT instead of OpenVLAOpenVLA 7B · Real-world in-distribution taskssuccess rate: 52% -> 72%Improved
train with HyT on top of VLA-OFT, Prismatic initializationVLA-OFT · LIBERO Goal and Longsuccess rateImproved
train with HyT instead of VLA-OFT starting from OpenVLAVLA-OFT · LIBERO (OpenVLA init)overall success: 95.3%No clear effect
train with HyT instead of standard VLA training at all data scalesVLA · ClevrSkillssuccess rateImproved
use ECoT-style chain-of-thought generation at inference instead of HyT act modeVLA · ClevrSkillsinference speed: 3x slowerHurt
use HiRobot hierarchical generation instead of HyT act modeVLA · ClevrSkillsinference speed: 4x slowerHurt
replace self-predicted thoughts with oracle thoughts at inferenceHyT · ClevrSkills complex taskssuccess rateImproved
add the think distribution to training coefficients (ablation)HyT · ClevrSkills (300 demos)success rateImproved
add the follow distribution to training coefficients (ablation)HyT · ClevrSkills oracle instruction followingsuccess rateImproved
add the follow distribution to training coefficients (acting mode alone) (ablation)HyT · ClevrSkills acting modesuccess rateNo clear effect

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