paper-with-me

Papers

Stabilizing Reinforcement Learning for Diffusion Language Models

2026-03-06 · Jianyuan Zhong, Kaibo Wang, Ding Ding, Zijin Feng, Haoli Bai, Yang Xiang, Jiacheng Sun, Qiang Xu arxiv

Group Relative Policy Optimization (GRPO) is highly effective for post-training autoregressive (AR) language models, yet its direct application to diffusion large language models (dLLMs) often triggers reward collapse. We identify two sources of incompatibility. First, GRPO relies on importance ratios defined by sequence probabilities, which are intractable in dLLMs and must be estimated (e.g., via ELBO-based or mean-field likelihood proxies), yielding inherently noisy ratios. Second, standard GRPO's formulation is not designed for estimated ratios: its conditional clipping can be anomalously bypassed by model-agnostic estimation noise, producing gradient spikes, while its fixed group-size normalization amplifies gradient-magnitude fluctuations under high-variance ratio estimates. We show these effects form a self-reinforcing instability loop that drives policy drift and further increases ratio variance. To break this loop, we propose StableDRL, a reformulation of GRPO tailored for dLLMs that uses (i) unconditional clipping to suppress outlier-induced spikes and (ii) self-normalization to constrain updates within the convex hull of per-sample gradients. We further extend StableDRL to block-wise diffusion models via a staircase attention mechanism.

📄 PDF Abstract BibTeX arXiv:2603.06743

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Maximum Entropy Inverse Reinforcement Learning of Diffusion Models with Energy-Based Models

2024-06-30 · Sangwoong Yoon, Himchan Hwang, Dohyun Kwon, Yung-Kyun Noh 외

We present a maximum entropy inverse reinforcement learning (IRL) approach for improving the sample quality of diffusion generative models, especially when the number of generation time steps is small. Similar to how IRL…

Anomaly Detection

Stabilizing MoE Reinforcement Learning by Aligning Training and Inference Routers

2025-10-13 · Wenhan Ma, Hailin Zhang, Liang Zhao, Yifan Song 외 arxiv

Reinforcement learning (RL) has emerged as a crucial approach for enhancing the capabilities of large language models. However, in Mixture-of-Experts (MoE) models, the routing mechanism often introduces instability, even…

Reinforcement Learning

Manifold-Guided Lyapunov Control with Diffusion Models

2024-03-26 · Amartya Mukherjee, Thanin Quartz, Jun Liu

This paper presents a novel approach to generating stabilizing controllers for a large class of dynamical systems using diffusion models. The core objective is to develop stabilizing control functions by identifying the …

A note on stabilizing reinforcement learning

2021-11-24 · Pavel Osinenko, Grigory Yaremenko, Ilya Osokin

Reinforcement learning is a general methodology of adaptive optimal control that has attracted much attention in various fields ranging from video game industry to robot manipulators. Despite its remarkable performance d…

reinforcement-learningReinforcement LearningReinforcement Learning (RL)

AIS: Adaptive Importance Sampling for Quantized RL

2026-05-13 · Jiajun Zhou, Wei Shao, Lingchao Zheng, Yuwei Fan 외 arxiv

Reinforcement learning (RL) for large language models (LLMs) is dominated by the cost of rollout generation, which has motivated the use of low-precision rollouts (e.g., FP8) paired with a BF16 trainer to improve through…

Reinforcement LearningMathematical Reasoning