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

• URL: http://arxiv.org/abs/2609.13679v1
• Date: Sat, 12 Sep 2026 03:12:46 GMT
• Title: How to Better Train VLAs: Lessons Learned From the REAL-I Challenge at ICRA 2026
• Abstract summary: 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.<n>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.<n>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.
• Score: 19.591059484557256
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• 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.

Related papers

• World-Action Models for Robot Learning and Control: A Survey [41.45329550202786]
  Recent advances in world models, video generation, and Vision-Language-Action (VLA) policies have motivated the development of World-Action Models (WAMs)<n>This survey provides a robotics-oriented review of WAMs.<n>We clarify their scope relative to conventional world models, model-based reinforcement learning, action-conditioned video generation, and reactive VLA policies, and organize existing methods through a unified taxonomy.<n>We discuss key challenges in action alignment, world-action factorization, spatial and multi-view consistency, long-horizon memory, neural simulation for closed-loop policy learning, and efficient inference
  arXiv  Detail & Related papers  (2026-09-13T16:35:56Z)
• One Demonstration Is Enough for Real-World Robotic Reinforcement Learning [34.23164234669014]
  We present AutoSERL, a framework that leverages a single demonstration to fully automate the intervention process in real-world robot RL.<n>We evaluate AutoSERL on six contact-intensive manipulation tasks across two robot platforms, spanning insertion, hanging, and hinge-based tasks.
  arXiv  Detail & Related papers  (2026-07-02T03:23:40Z)
• Two Bridges, One Pathway: From VLMs to Generalizable VLAs with Embodied Trajectory-Coupled Data [93.7685703383343]
  Vision-language models (VLMs) are powerful general-purpose reasoners, yet converting them into robot control policies is surprisingly difficult.<n>We argue that this gap can be bridged gradually with the right intermediate data.<n>We show that mixing task-relevant out-of-distribution ETC data with a small amount of action data enables the model to generalize to novel visual-language conditions.
  arXiv  Detail & Related papers  (2026-06-07T08:57:51Z)
• Rethinking Visual-Language-Action Model Scaling: Alignment, Mixture, and Regularization [65.37179698521766]
  Vision-Language-Action (VLA) models show strong promise for generalist robot control.<n>It remains unclear whether -- and under what conditions -- the standard "scale data" recipe translates to robotics.<n>We present a systematic, controlled study of VLA scaling that revisits core training choices for pretraining across diverse robots.
  arXiv  Detail & Related papers  (2026-02-10T12:25:43Z)
• A Systematic Study of Data Modalities and Strategies for Co-training Large Behavior Models for Robot Manipulation [11.026552246133521]
  Large behavior models have shown strong dexterous manipulation capabilities by extending imitation learning to large-scale training on multi-task robot data.<n>Recent work relies on co-training: jointly learning from target robot data and heterogeneous data modalities.<n>We present a large-scale empirical study examining five co-training data modalities: standard vision-language data, dense language annotations for robot trajectories, cross-embodiment robot data, human videos, and discrete robot action tokens.
  arXiv  Detail & Related papers  (2026-02-01T07:22:55Z)
• Guardian: Detecting Robotic Planning and Execution Errors with Vision-Language Models [53.20969621498248]
  We propose an automatic robot failure synthesis approach that procedurally perturbs successful trajectories to generate diverse planning and execution failures.<n>We construct three new failure detection benchmarks: RLBench-Fail, BridgeDataV2-Fail, and UR5-Fail.<n>We then train Guardian, a VLM with multi-view images for detailed failure reasoning and detection.
  arXiv  Detail & Related papers  (2025-12-01T17:57:27Z)
• Action Flow Matching for Continual Robot Learning [54.10050120844738]
  Continual learning in robotics seeks systems that can constantly adapt to changing environments and tasks.<n>We introduce a generative framework leveraging flow matching for online robot dynamics model alignment.<n>We find that by transforming the actions themselves rather than exploring with a misaligned model, the robot collects informative data more efficiently.
  arXiv  Detail & Related papers  (2025-04-25T16:26:15Z)
• VITAL: Interactive Few-Shot Imitation Learning via Visual Human-in-the-Loop Corrections [10.49712834719005]
  Imitation Learning (IL) has emerged as a powerful approach in robotics, allowing robots to acquire new skills by mimicking human actions.<n>Despite its potential, the data collection process for IL remains a significant challenge due to logistical difficulties and high costs associated with obtaining high-quality demonstrations.<n>We propose a large-scale data generation from a handful of demonstrations through data augmentation in simulation.
  arXiv  Detail & Related papers  (2024-07-30T23:29:47Z)
• Interactive Planning Using Large Language Models for Partially Observable Robotics Tasks [54.60571399091711]
  Large Language Models (LLMs) have achieved impressive results in creating robotic agents for performing open vocabulary tasks. We present an interactive planning technique for partially observable tasks using LLMs.
  arXiv  Detail & Related papers  (2023-12-11T22:54:44Z)
• Robot Fine-Tuning Made Easy: Pre-Training Rewards and Policies for Autonomous Real-World Reinforcement Learning [58.3994826169858]
  We introduce RoboFuME, a reset-free fine-tuning system for robotic reinforcement learning. Our insights are to utilize offline reinforcement learning techniques to ensure efficient online fine-tuning of a pre-trained policy. Our method can incorporate data from an existing robot dataset and improve on a target task within as little as 3 hours of autonomous real-world experience.
  arXiv  Detail & Related papers  (2023-10-23T17:50:08Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.