Beyond Imitation: Reinforcement Learning–Based Sim–Real Co-Training for VLA Models

Liangzhi Shi, Shuaihang Chen, Feng Gao, Yinuo Chen, Kang Chen, Tonghe Zhang, Hongzhi Zang, Weinan Zhang, Chao Yu, Yu Wang

Keywords: Reinforcement Learning; Sim-Real Co-Training; Supervised Fine-tuning

TL;DR: We introduce RL-based sim–real co-training for VLA models, leveraging interactive simulation with real-data regularization to outperform SFT-based methods in real-world success, generalization, and data efficiency.

Abstract: Simulation offers a scalable and low-cost way to enrich vision-language-action (VLA) training, reducing reliance on expensive real-robot demonstrations. However, most sim-real co-training methods rely on supervised fine-tuning (SFT), which treats simulation as a static source of demonstrations and does not exploit large-scale closed-loop interaction. Consequently, real-world gains and generalization are often limited. In this paper, we propose an \underline{\textit{RL}}-based sim-real \underline{\textit{Co}}-training \modify{(RL-Co)} framework that leverages interactive simulation while preserving real-world capabilities. Our method follows a generic two-stage design: we first warm-start the policy with SFT on a mixture of real and simulated demonstrations, then fine-tune it with reinforcement learning in simulation while adding an auxiliary supervised loss on real-world data to anchor the policy and mitigate catastrophic forgetting. We evaluate our framework on four real-world tabletop manipulation tasks using two representative VLA architectures, OpenVLA and $\pi_{0.5}$, and observe consistent improvements over real-only fine-tuning and SFT-based co-training, including +24% real-world success on OpenVLA and +20% on $\pi_{0.5}$. Beyond higher success rates, RL co-training yields stronger generalization to unseen task variations and substantially improved real-world data efficiency, providing a practical and scalable pathway for leveraging simulation to enhance real-robot deployment.

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 3