Skip to content
Preprint

How to Better Train VLAs: Lessons Learned From the REAL-I Challenge at ICRA 2026

Sep 2026 · 0 citations · 13 references
Computer Science

TL;DR

The team reports highlight the importance of adapting to the deployment environment while retaining prior capabilities, treating demonstration quality at an appropriate temporal scale, and suppressing errors in inactive robot components, and expose the limitations of offline action-prediction metrics for forecasting closed-loop success.

Abstract

How can robot policies learn more effectively from a fixed demonstration budget? The first Real-world Embodied AI Learning (REAL-I) Challenge at ICRA 2026 examined this question through simulation, real-robot evaluation, and an on-site final on a shared dual-arm humanoid platform. We describe the challenge tasks, data and deployment interfaces, and competition results, then compare the approaches contributed by NUS-CLEAR, RCL-Lab, and DeepTouch AI. Their systems combined pretrained vision-language-action models and task-specific imitation policies with different strategies for data curation, staged adaptation, checkpoint selection, and action-space design. The team reports highlight the importance of adapting to the deployment environment while retaining prior capabilities, treating demonstration quality at an appropriate temporal scale, and suppressing errors in inactive robot components. They also expose the limitations of offline action-prediction metrics for forecasting closed-loop success. These observations motivate a view of fixed-data robot learning that integrates data, adaptation, evaluation, and deployment.

View source

Similar papers

Sep 2026

Find Something You Can't Do: Agentic Real-World Reinforcement Learning for Self-Improving VLA Models

FIND is introduced, an agentic real-world RL framework that closes the loop between scene understanding, weakness-aware practice, self-evaluation, and policy improvement in a persistent workspace and reframes autonomous practice as a scene-conditioned, performance-aware task-selection problem.

Yuan Fang, Ze-Chu Li, Hao-Lei Tong et al. · 4 citations · ⚡1
#artificial intelligence Preprint Sep 2026

ReWeight: Leveraging Human Data for VLA Post-Training via Demonstration Retrieval and Sample Weighting

ReWeight is introduced, a framework that incorporates human data into VLA post-training through demonstration-level retrieval and sample-level weighting and learns a cross-embodiment visuomotor representation that combines visual observations with future actions to measure behavioral similarity between human and robot...

Chen-Wei Wang, Dian-Ye Huang, Match W. L. Ko et al. · 0 citations
#machine learning Preprint Oct 2026

How (and How Not) to Use Data Augmentation in VLA Post-Training

Vision-language-action (VLA) models currently demonstrate strong performance in a wide range of real-world robotics tasks. However, they often still lack the generalization ability to handle large visual out-of-distribution shifts. Post-training of VLAs with reinforcement learning (RL) has been shown to benefit robustn...

Bram Grooten, J. Vanschoren · 0 citations
#machine learning Preprint Oct 2026

NestedVLA: Learning to Consolidate and Generate Skills for Vision-Language-Action Models

Vision-language-action (VLA) models can learn a wide range of robotic skills, but combining separately trained skills remains challenging. The original training data may be unavailable for joint retraining, while directly merging model weights can cause interference. We propose NestedVLA, which consolidates independent...

Guo-Dong Du, Wan-Yu Lin · 1 citation

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.