# CSO: Refining Robotic Policies via Skill Distribution Alignment and Skill-Grained Optimization

Preprint · Zhiyuan Xiang, Xiang Deng, Wanjie Tao, Xu Chen, Zhao Jielun, Weili Guan

- Archive id: arxiv:2610.38357
- Published: 2026-10-28
- arXiv: https://arxiv.org/abs/2610.38357
- Source URL: https://arxiv.org/abs/2610.38357
- PDF: https://arxiv.org/pdf/2610.38357
- Provenance: arxiv

## Abstract

Discretizing continuous actions into discrete skills using methods like VQ-VAE has emerged as a powerful paradigm for robotic manipulation. However, quantization errors yield a suboptimal distribution for the skill prior, degrading its performance. While reinforcement learning offers a path for refinement, its direct application faces unstable encoder updates and a granularity dilemma in importance sampling. To address these challenges, we introduce Cascaded Skills Optimization (CSO), a two-stage post-training framework. First, Rejection-Sampling Supervised Fine-tuning aligns the model's observation-to-skill mapping with the distribution of successful online trajectories via supervised fine-tuning. Second, with the observation encoder frozen, Skills Policy Optimization computes an independent, clipped importance ratio for each complete skill, enabling stable and efficient updates. With a lightweight QueST-style policy, CSO raises overall LIBERO success from 80.1% to 97.5%, improves performance on RoboTwin2.0, and achieves 61.0% success across five physical single-arm tasks. Additional comparisons with FAST, discrete BEAST, and fixed-buffer IQL further validate the effectiveness of our two-stage post-training strategy.

_Abstract reproduced under CC0-1.0 (arXiv descriptive metadata)._

Topics: [Efficient Training & Inference](https://space-kitty.com/topics/efficient-ai)

---
Space Kitty Archive · https://space-kitty.com · API: https://space-kitty.com/api/v1 · MCP: https://space-kitty.com/mcp · Premium: $0.20/request via x402 (Base Sepolia (testnet)). Metadata: arXiv (CC0), OpenAlex (CC0), Europe PMC (CC BY/CC0 works), ClinicalTrials.gov.