One-Token Rollout: Guiding Supervised Fine-Tuning of LLMs with Policy Gradient

• URL: http://arxiv.org/abs/2509.26313v1
• Date: Tue, 30 Sep 2025 14:25:56 GMT
• Title: One-Token Rollout: Guiding Supervised Fine-Tuning of LLMs with Policy Gradient
• Authors: Rui Ming, Haoyuan Wu, Shoubo Hu, Zhuolun He, Bei Yu,
• Abstract summary: We introduce one-token rollout (OTR), a novel fine-tuning algorithm that guides SFT with the policy gradient method.<n>OTR reframes the autoregressive learning process by treating each token generation as a single-step reinforcement learning trajectory.<n>Our findings establish OTR as a powerful and practical alternative for fine-tuning LLMs.
• Score: 16.05489579792086
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Supervised fine-tuning (SFT) is the predominant method for adapting large language models (LLMs), yet it often struggles with generalization compared to reinforcement learning (RL). In this work, we posit that this performance disparity stems not just from the loss function, but from a more fundamental difference: SFT learns from a fixed, pre-collected dataset, whereas RL utilizes on-policy data sampled from the current policy. Building on this hypothesis, we introduce one-token rollout (OTR), a novel fine-tuning algorithm that guides SFT with the policy gradient method. OTR reframes the autoregressive learning process by treating each token generation as a single-step reinforcement learning trajectory. At each step, it performs a Monte Carlo ``rollout'' by sampling multiple candidate tokens from the current policy's distribution. The ground-truth token from the supervised data is then used to provide a reward signal to these samples. Guided by policy gradient, our algorithm repurposes static, off-policy supervised data into a dynamic, on-policy signal at the token level, capturing the generalization benefits of on-policy learning while bypassing the costly overhead of full sentence generation. Through extensive experiments on a diverse suite of challenging benchmarks spanning mathematical reasoning, code generation, and general domain reasoning, we demonstrate that OTR consistently outperforms standard SFT. Our findings establish OTR as a powerful and practical alternative for fine-tuning LLMs and provide compelling evidence that the on-policy nature of data is a critical driver of generalization, offering a promising new direction for fine-tuning LLMs.

Related papers

• Retaining by Doing: The Role of On-Policy Data in Mitigating Forgetting [40.80967570661867]
  Adapting language models to new tasks via post-training carries the risk of degrading existing capabilities.<n>We compare the forgetting patterns of two widely adopted post-training methods: supervised fine-tuning (SFT) and reinforcement learning (RL)<n>RL leads to less forgetting than SFT while achieving comparable or higher target task performance.
  arXiv  Detail & Related papers  (2025-10-21T17:59:41Z)
• Beyond Static LLM Policies: Imitation-Enhanced Reinforcement Learning for Recommendation [23.945049006150555]
  Large language models (LLMs) have become critical tools for enhancing user engagement by delivering personalized content across diverse digital platforms.<n>Direct deployment of LLMs as primary recommendation policies presents notable challenges, including persistent latency issues.<n>This paper proposes a novel offline reinforcement learning framework that leverages imitation learning from LLM-generated trajectories.
  arXiv  Detail & Related papers  (2025-10-15T07:28:29Z)
• Group-Relative REINFORCE Is Secretly an Off-Policy Algorithm: Demystifying Some Myths About GRPO and Its Friends [64.71326476563213]
  Off-policy reinforcement learning for large language models (LLMs) is attracting growing interest.<n>We present a first-principles derivation for grouprelative REINFORCE without assuming a specific training data distribution.<n>This perspective yields two general principles for adapting REINFORCE to off-policy settings.
  arXiv  Detail & Related papers  (2025-09-29T02:34:54Z)
• Contrastive Policy Gradient: Aligning LLMs on sequence-level scores in a supervised-friendly fashion [43.77763433288893]
  We introduce Contrastive Policy Gradient, or CoPG, a simple and mathematically principled new RL algorithm that can estimate the optimal policy even from off-policy data.<n>We show this approach to generalize the direct alignment method IPO (identity preference optimization) and classic policy gradient.<n>We experiment with the proposed CoPG on a toy bandit problem to illustrate its properties, as well as for finetuning LLMs on a summarization task.
  arXiv  Detail & Related papers  (2024-06-27T14:03:49Z)
• Fine-Tuning Language Models with Reward Learning on Policy [68.70065254564642]
  Reinforcement learning from human feedback (RLHF) has emerged as an effective approach to aligning large language models (LLMs) to human preferences. Despite its popularity, (fixed) reward models may suffer from inaccurate off-distribution. We propose reward learning on policy (RLP), an unsupervised framework that refines a reward model using policy samples to keep it on-distribution.
  arXiv  Detail & Related papers  (2024-03-28T10:02:10Z)
• How Can LLM Guide RL? A Value-Based Approach [68.55316627400683]
  Reinforcement learning (RL) has become the de facto standard practice for sequential decision-making problems by improving future acting policies with feedback. Recent developments in large language models (LLMs) have showcased impressive capabilities in language understanding and generation, yet they fall short in exploration and self-improvement capabilities. We develop an algorithm named LINVIT that incorporates LLM guidance as a regularization factor in value-based RL, leading to significant reductions in the amount of data needed for learning.
  arXiv  Detail & Related papers  (2024-02-25T20:07:13Z)
• Nash Learning from Human Feedback [86.09617990412941]
  We introduce an alternative pipeline for the fine-tuning of large language models using pairwise human feedback. We term this approach Nash learning from human feedback (NLHF) We present a novel algorithmic solution, Nash-MD, founded on the principles of mirror descent.
  arXiv  Detail & Related papers  (2023-12-01T19:26:23Z)
• Iteratively Refined Behavior Regularization for Offline Reinforcement Learning [57.10922880400715]
  In this paper, we propose a new algorithm that substantially enhances behavior-regularization based on conservative policy iteration. By iteratively refining the reference policy used for behavior regularization, conservative policy update guarantees gradually improvement. Experimental results on the D4RL benchmark indicate that our method outperforms previous state-of-the-art baselines in most tasks.
  arXiv  Detail & Related papers  (2023-06-09T07:46:24Z)
• Jump-Start Reinforcement Learning [68.82380421479675]
  We present a meta algorithm that can use offline data, demonstrations, or a pre-existing policy to initialize an RL policy. In particular, we propose Jump-Start Reinforcement Learning (JSRL), an algorithm that employs two policies to solve tasks. We show via experiments that JSRL is able to significantly outperform existing imitation and reinforcement learning algorithms.
  arXiv  Detail & Related papers  (2022-04-05T17:25:22Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.