A Critical Look At Tokenwise Reward-Guided Text Generation

• URL: http://arxiv.org/abs/2406.07780v1
• Date: Wed, 12 Jun 2024 00:19:40 GMT
• Title: A Critical Look At Tokenwise Reward-Guided Text Generation
• Authors: Ahmad Rashid, Ruotian Wu, Julia Grosse, Agustinus Kristiadi, Pascal Poupart,
• Abstract summary: We show that reward models trained on full sequences are not compatible with scoring partial sequences. We propose to explicitly train a Bradley-Terry reward model on partial sequences, and autoregressively sample from the implied tokenwise policy during decoding time.
• Score: 23.908449840589284
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models (LLMs) can significantly be improved by aligning to human preferences -- the so-called reinforcement learning from human feedback (RLHF). However, the cost of fine-tuning an LLM is prohibitive for many users. Due to their ability to bypass LLM finetuning, tokenwise reward-guided text generation (RGTG) methods have recently been proposed. They use a reward model trained on full sequences to score partial sequences during a tokenwise decoding, in a bid to steer the generation towards sequences with high rewards. However, these methods have so far been only heuristically motivated and poorly analyzed. In this work, we show that reward models trained on full sequences are not compatible with scoring partial sequences. To alleviate this issue, we propose to explicitly train a Bradley-Terry reward model on partial sequences, and autoregressively sample from the implied tokenwise policy during decoding time. We study the property of this reward model and the implied policy. In particular, we show that this policy is proportional to the ratio of two distinct RLHF policies. We show that our simple approach outperforms previous RGTG methods and achieves similar performance as strong offline baselines but without large-scale LLM finetuning.

Related papers

• Towards Cost-Effective Reward Guided Text Generation [27.11836864643437]
  Reward-guided text generation (RGTG) has emerged as a viable alternative to offline reinforcement learning from human feedback. We present a novel reward model architecture that is trained, using a Bradley-Terry loss, to prefer the optimal expansion of a sequence with just a emphsingle call to the reward model at each step of the generation process.
  arXiv  Detail & Related papers  (2025-02-06T21:36:44Z)
• Segmenting Text and Learning Their Rewards for Improved RLHF in Language Model [96.20350225621813]
  Reinforcement learning from human feedback (RLHF) has been widely adopted to align language models (LMs) with human preference. In this paper, we seek to get the best of both by training and utilizing a segment-level reward model.
  arXiv  Detail & Related papers  (2025-01-06T06:17:56Z)
• T-REG: Preference Optimization with Token-Level Reward Regularization [35.07328450591201]
  Reinforcement learning from human feedback (RLHF) has been crucial in aligning large language models with human values. Recent methods have attempted to address this limitation by introducing token-level rewards. We propose token-level reward regularization (T-REG), a novel approach that leverages both sequence-level and token-level rewards for preference optimization.
  arXiv  Detail & Related papers  (2024-12-03T18:56:07Z)
• Optimal Design for Reward Modeling in RLHF [83.3614658277817]
  We formalize the reward training model in Reinforcement Learning from Human Feedback. We frame the selection of an effective dataset as a simple regret minimization task. We derive bounds on the simple regret under appropriate assumptions.
  arXiv  Detail & Related papers  (2024-10-22T14:36:44Z)
• 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. We show this approach to generalize the direct alignment method IPO (identity preference optimization) and classic policy gradient. 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)
• Dense Reward for Free in Reinforcement Learning from Human Feedback [64.92448888346125]
  We leverage the fact that the reward model contains more information than just its scalar output. We use these attention weights to redistribute the reward along the whole completion. Empirically, we show that it stabilises training, accelerates the rate of learning, and, in practical cases, may lead to better local optima.
  arXiv  Detail & Related papers  (2024-02-01T17:10:35Z)
• 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)
• Text Generation with Efficient (Soft) Q-Learning [91.47743595382758]
  Reinforcement learning (RL) offers a more flexible solution by allowing users to plug in arbitrary task metrics as reward. We introduce a new RL formulation for text generation from the soft Q-learning perspective. We apply the approach to a wide range of tasks, including learning from noisy/negative examples, adversarial attacks, and prompt generation.
  arXiv  Detail & Related papers  (2021-06-14T18:48:40Z)

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.