SRPO: Self-Referential Policy Optimization for Vision-Language-Action Models

TL;DR

SRPO introduces a self-referential reinforcement learning framework for vision-language-action models that leverages the model's own successful trajectories and latent world representations to improve robotic manipulation performance efficiently.

Contribution

It proposes a novel RL framework that eliminates the need for external demonstrations by using self-generated trajectories and latent space representations for progress measurement.

Findings

Achieves 99.2% success rate on LIBERO benchmark in 200 steps.

Outperforms previous methods with a 103% success rate increase.

Demonstrates robustness with 167% improvement on LIBERO-Plus.

Abstract

Vision-Language-Action (VLA) models excel in robotic manipulation but are constrained by their heavy reliance on expert demonstrations, leading to demonstration bias and limiting performance. Reinforcement learning (RL) is a vital post-training strategy to overcome these limits, yet current VLA-RL methods, including group-based optimization approaches, are crippled by severe reward sparsity. Relying on binary success indicators wastes valuable information in failed trajectories, resulting in low training efficiency. To solve this, we propose Self-Referential Policy Optimization (SRPO), a novel VLA-RL framework. SRPO eliminates the need for external demonstrations or manual reward engineering by leveraging the model's own successful trajectories, generated within the current training batch, as a self-reference. This allows us to assign a progress-wise reward to failed attempts. A core innovation is the use of latent world representations to measure behavioral progress robustly. Instead of relying on raw pixels or requiring domain-specific fine-tuning, we utilize the compressed, transferable encodings from a world model's latent space. These representations naturally capture progress patterns across environments, enabling accurate, generalized trajectory comparison. Empirical evaluations on the LIBERO benchmark demonstrate SRPO's efficiency and effectiveness. Starting from a supervised baseline with 48.9% success, SRPO achieves a new state-of-the-art success rate of 99.2% in just 200 RL steps, representing a 103% relative improvement without any extra supervision. Furthermore, SRPO shows substantial robustness, achieving a 167% performance improvement on the LIBERO-Plus benchmark.