Reinforcement Learning (RL) has proven highly effective for autoregressive language models, but adapting these methods to diffusion large language models (dLLMs) presents fundamental challenges. The core difficulty lies in likelihood approximation: while autoregressive models naturally provide token-level conditional probabilities essential for token-level RL objectives (e.g., GRPO), dLLMs generate sequences through iterative non-autoregressive denoising steps that lack this factorization. To address this fundamental mismatch, we propose ELBO-based Sequence-level Policy Optimization (ESPO), a principled RL framework that treats entire sequence generation as a single action and uses the ELBO as a tractable sequence-level likelihood proxy. Our method incorporates per-token normalization of importance ratios and robust KL-divergence estimation to ensure stable large-scale training. Extensive experiments on mathematical reasoning, coding, and planning tasks demonstrate that ESPO significantly outperforms token-level baselines, achieving dramatic improvements of 20-40 points on the Countdown task, while maintaining consistent gains on math and coding benchmarks. Our approach establishes sequence-level optimization as a principled and empirically effective paradigm for RL in dLLMs. Our code is available at https://github.com/ML-GSAI/ESPO. Large language models (LLMs) (OpenAI, 2023) have become a cornerstone of modern natural language processing, achieving remarkable progress across math (Guo et al., 2025), coding (Hui et al., 2024), and planning tasks (Y ao et al., 2023). While autoregressive (AR) modeling has long dominated this field, recent advances in diffusion large language models (dLLMs) have demonstrated strong potential as an alternative formulation (Ou et al., 2024; Shi et al., 2024; Sahoo et al., 2024; Nie et al., 2025; Y e et al., 2025). With the advent of powerful pretrained dLLMs, the next frontier lies in post-training (Ouyang et al., 2022) to further enhance their capabilities. Among various post-training paradigms, reinforcement learning (RL) has emerged as a powerful approach that enables test-time scaling (Snell et al., 2025) through verifiable rewards (Guo et al., 2025). It has yielded substantial gains on reasoning tasks in recent AR models (OpenAI, 2024), such as math (Cobbe et al., 2021b), coding (Chen et al., 2021), and reasoning (Liu et al., 2023b).

Large language model (LLM) reinforcement learning has increasingly relied on group-based policy optimization frameworks, such as GRPO and GSPO, to achieve stable fine-tuning at scale. However, a fundamental trade-off persists between optimization granularity and training stability. While GSPO improves robustness via sequence-level optimization, its monolithic treatment of sequences introduces severe inefficiencies: its conservative clipping mechanism indiscriminately discards valid training samples-a phenomenon we term gradient underutilization-and its uniform credit assignment fails to capture the heterogeneous contributions of critical reasoning steps. In this work, we propose Entropy Importance Sampling Policy Optimization (ESPO), a novel framework that reconciles fine-grained control with training stability. ESPO decomposes sequences into groups based on predictive entropy, enabling (1) Entropy-driven Importance Sampling to capture intra-sequence heterogeneity, and (2) Entropy-adaptive Clipping to dynamically allocate trust regions based on model uncertainty. Extensive experiments on mathematical reasoning benchmarks demonstrate that ESPO not only accelerates convergence but also achieves state-of-the-art performance, notably improving accuracy on the challenging HMMT benchmark from 4.4% to 13.13%.

However, safe RL often suffers from sample inefficiency, requiring extensive interactions with the environment to learn a safe policy.

Safe reinforcement learning (RL) is crucial for deploying RL agents in real-world applications, as it aims to maximize long-term rewards while satisfying safety constraints. However, safe RL often suffers from sample inefficiency, requiring extensive interactions with the environment to learn a safe policy. We propose Efficient Safe Policy Optimization (ESPO), a novel approach that enhances the efficiency of safe RL through sample manipulation. ESPO employs an optimization framework with three modes: maximizing rewards, minimizing costs, and balancing the trade-off between the two. By dynamically adjusting the sampling process based on the observed conflict between reward and safety gradients, ESPO theoretically guarantees convergence, optimization stability, and improved sample complexity bounds. Experiments on the Safety-MuJoCo and Omnisafe benchmarks demonstrate that ESPO significantly outperforms existing primal-based and primal-dual-based baselines in terms of reward maximization and constraint satisfaction. Moreover, ESPO achieves substantial gains in sample efficiency, requiring 25--29% fewer samples than baselines, and reduces training time by 21--38%.

Recent success in Deep Reinforcement Learning (DRL) methods has shown that policy optimization with respect to an off-policy distribution via importance sampling is effective for sample reuse. In this paper, we show that the use of importance sampling could introduce high variance in the objective estimate. Specifically, we show in a principled way that the variance of importance sampling estimate grows quadratically with importance ratios and the large ratios could consequently jeopardize the effectiveness of surrogate objective optimization. We then propose a technique called sample dropout to bound the estimation variance by dropping out samples when their ratio deviation is too high. We instantiate this sample dropout technique on representative policy optimization algorithms, including TRPO, PPO, and ESPO, and demonstrate that it consistently boosts the performance of those DRL algorithms on both continuous and discrete action controls, including MuJoCo, DMControl and Atari video games. Our code is open-sourced at \url{https://github.com/LinZichuan/sdpo.git}.

Proximal Policy Optimization (PPO) methods learn a policy by iteratively performing multiple mini-batch optimization epochs of a surrogate objective with one set of sampled data. Ratio clipping PPO is a popular variant that clips the probability ratios between the target policy and the policy used to collect samples. Ratio clipping yields a pessimistic estimate of the original surrogate objective, and has been shown to be crucial for strong performance. We show in this paper that such ratio clipping may not be a good option as it can fail to effectively bound the ratios. Instead, one can directly optimize the original surrogate objective for multiple epochs; the key is to find a proper condition to early stop the optimization epoch in each iteration. Our theoretical analysis sheds light on how to determine when to stop the optimization epoch, and call the resulting algorithm Early Stopping Policy Optimization (ESPO). We compare ESPO with PPO across many continuous control tasks and show that ESPO significantly outperforms PPO. Furthermore, we show that ESPO can be easily scaled up to distributed training with many workers, delivering strong performance as well.