paper-with-me

Papers

Simple Policy Gradients for Reasoning with Diffusion Language Models

2025-10-05 · Anthony Zhan arxiv

Diffusion large language models (dLLMs) represent a promising alternative to autoregressive LLMs; however, the lack of effective post-training techniques, including reinforcement learning (RL), remains a key challenge for dLLMs, especially for downstream applications. Existing approaches often rely on a sequence-level view that requires biased likelihood approximations. In this work, we propose Amortized Group Relative Policy Optimization (AGRPO), a policy gradient algorithm that leverages the Markovian nature of dLLMs, optimizing individual denoising steps rather than full sequences. Our approach improves alignment between the trained policy and the inference process and also admits efficient, unbiased gradient updates via a novel timestep estimation scheme. We demonstrate AGRPO's effectiveness on different math and reasoning tasks, achieving absolute accuracy gains of +59.4\% and +69.7\% on Countdown and Sudoku over the base LLaDA model, exceeding comparable methods such as diffu-GRPO. Code is available at https://github.com/probablyabot/agrpo.

📄 PDF Abstract BibTeX arXiv:2510.04019

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Mask-Aware Policy Gradients for Diffusion Language Models

2026-07-16 · Haran Raajesh, Kulin Shah, Adam Klivans, Philipp Krähenbühl arxiv

Reinforcement learning has proven effective for improving reasoning in large language models, but extending it to Masked Diffusion Language Models (MDLMs) remains challenging due to the intractability of the log-likeliho…

Mathematical ReasoningReinforcement Learning

Reasoning in Diffusion Large Language Models is Concentrated in Dynamic Confusion Zones

2025-11-19 · Ranfei Chen, Ming Chen, Kaifei Wang arxiv

Diffusion Large Language Models (dLLMs) are rapidly emerging alongside autoregressive models as a powerful paradigm for complex reasoning, with reinforcement learning increasingly used for downstream alignment. Existing …

Reinforcement Learning

Inpainting-Guided Policy Optimization for Diffusion Large Language Models

2025-09-12 · Siyan Zhao, Mengchen Liu, Jing Huang, Miao Liu 외 arxiv

Masked diffusion large language models (dLLMs) are emerging as promising alternatives to autoregressive LLMs, offering competitive performance while supporting unique generation capabilities such as inpainting. We explor…

Reinforcement Learning

Beyond Token-Level Policy Gradients for Complex Reasoning with Large Language Models

2026-02-16 · Mufan Xu, Kehai Chen, Xuefeng Bai, Zhengyu Niu 외 arxiv

Existing policy-gradient methods for auto-regressive language models typically select subsequent tokens one at a time as actions in the policy. While effective for many generation tasks, such an approach may not fully ca…

Mathematical Reasoning

Back on Track: Aligning Rewards and States for Reasoning in Diffusion Large Language Models

2026-06-07 · Yawen Shao, Jie Xiao, Kai Zhu, Yu Liu 외 arxiv

Reinforcement learning (RL) holds immense promise for enhancing the reasoning capabilities of diffusion large language models (dLLMs). However, progress is fundamentally constrained by a dual misalignment between authent…

Reinforcement Learning