# Greedy Sampling Is Provably Efficient for RLHF

arXiv: https://arxiv.org/abs/2510.24700 · HTML: https://mlsanity.com/paper/2510.24700/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Placed in Reward modeling and RLHF, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 45 of its 59 references that are in the map.
- First posted on arXiv 28 Oct 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Reward modeling and RLHF](https://mlsanity.com/alignment/reward-modeling-and-rlhf/)

## Builds on (most cited first, 30 of 45)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Learning to summarize from human feedback](https://mlsanity.com/paper/2009.01325/) (2020)
- [Fine-Tuning Language Models from Human Preferences](https://mlsanity.com/paper/1909.08593/) (2019)
- [A General Theoretical Paradigm to Understand Learning from Human Preferences](https://mlsanity.com/paper/2310.12036/) (2023)
- [RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment](https://mlsanity.com/paper/2304.06767/) (2023)
- [Provably Efficient Reinforcement Learning with Linear Function Approximation](https://mlsanity.com/paper/1907.05388/) (2019)
- [Minimax Regret Bounds for Reinforcement Learning](https://mlsanity.com/paper/1703.05449/) (2017)
- [Is Q-learning Provably Efficient?](https://mlsanity.com/paper/1807.03765/) (2018)
- [SLiC-HF: Sequence Likelihood Calibration with Human Feedback](https://mlsanity.com/paper/2305.10425/) (2023)
- [Is Pessimism Provably Efficient for Offline RL?](https://mlsanity.com/paper/2012.15085/) (2020)
- [Iterative Preference Learning from Human Feedback: Bridging Theory and Practice for RLHF under KL-Constraint](https://mlsanity.com/paper/2312.11456/) (2023)
- [Bridging Offline Reinforcement Learning and Imitation Learning: A Tale of Pessimism](https://mlsanity.com/paper/2103.12021/) (2021)
- [Bellman-consistent Pessimism for Offline Reinforcement Learning](https://mlsanity.com/paper/2106.06926/) (2021)
- [Direct Language Model Alignment from Online AI Feedback](https://mlsanity.com/paper/2402.04792/) (2024)
- [Principled Reinforcement Learning with Human Feedback from Pairwise or K-wise Comparisons](https://mlsanity.com/paper/2301.11270/) (2023)
- [From r to Q^*: Your Language Model is Secretly a Q-Function](https://mlsanity.com/paper/2404.12358/) (2024)
- [Nash Learning from Human Feedback](https://mlsanity.com/paper/2312.00886/) (2023)
- [Provably Efficient Exploration in Policy Optimization](https://mlsanity.com/paper/1912.05830/) (2019)
- [Policy Finetuning: Bridging Sample-Efficient Offline and Online Reinforcement Learning](https://mlsanity.com/paper/2106.04895/) (2021)
- [DPO Meets PPO: Reinforced Token Optimization for RLHF](https://mlsanity.com/paper/2404.18922/) (2024)
- [Dueling RL: Reinforcement Learning with Trajectory Preferences](https://mlsanity.com/paper/2111.04850/) (2021)
- [Exploratory Preference Optimization: Harnessing Implicit Q*-Approximation for Sample-Efficient RLHF](https://mlsanity.com/paper/2405.21046/) (2024)
- [Contextual Dueling Bandits](https://mlsanity.com/paper/1502.06362/) (2015)
- [Is RLHF More Difficult than Standard RL?](https://mlsanity.com/paper/2306.14111/) (2023)
- [Settling the Sample Complexity of Model-Based Offline Reinforcement Learning](https://mlsanity.com/paper/2204.05275/) (2022)

## Built on by (most cited first, 11 of 11)

- [Provably Efficient Regularized Online RLHF with Generalized Bilinear Preferences](https://mlsanity.com/paper/2602.23116/) (2026)
- [Near-Optimal Regret for KL-Regularized Multi-Armed Bandits](https://mlsanity.com/paper/2603.02155/) (2026)
- [On the Optimal Sample Complexity of Offline Multi-Armed Bandits with KL Regularization](https://mlsanity.com/paper/2605.02141/) (2026)
- [Fast Rates for Offline Contextual Bandits with Forward-KL Regularization under Single-Policy Concentrability](https://mlsanity.com/paper/2605.09214/) (2026)
- [Online KL-Regularized Reinforcement Learning with Function Approximation under Misspecification](https://mlsanity.com/paper/2606.06053/) (2026)
- [f-Divergence Regularized RLHF: Two Tales of Sampling and Unified Analyses](https://mlsanity.com/paper/2605.06977/) (2026)
- [When Greedy Sampling Explores: KL-Regularized Contextual Bandits without Eluder-Dimension Dependence](https://mlsanity.com/paper/2609.13564/) (2026)
- [Demystifying the unreasonable effectiveness of online alignment methods](https://mlsanity.com/paper/2604.17207/) (2026)
- [Personalized Alignment Revisited: The Necessity and Sufficiency of User Diversity](https://mlsanity.com/paper/2605.09119/) (2026)
- [Efficient Exploration for Iterative Nash Preference Optimization](https://mlsanity.com/paper/2606.01382/) (2026)
- [Fast Rates for Offline Q-Learning with KL Regularization](https://mlsanity.com/paper/2610.18839/) (2026)

## Abstract (arXiv)

Reinforcement Learning from Human Feedback (RLHF) has emerged as a key technique for post-training large language models. Despite its empirical success, the theoretical understanding of RLHF is still limited, as learning the KL-regularized target with only preference feedback poses additional challenges compared with canonical RL. Existing works mostly study the reward-based Bradley-Terry (BT) preference model, and extend classical designs utilizing optimism or pessimism. This work, instead, considers the general preference model (whose practical relevance has been observed recently) and obtains performance guarantees with major, order-wise improvements over existing ones. Surprisingly, these results are derived from algorithms that directly use the empirical estimates (i.e., greedy sampling), as opposed to constructing optimistic or pessimistic estimates in previous works. This insight has a deep root in the unique structural property of the optimal policy class under the KL-regularized target, and we further specialize it to the BT model, highlighting the surprising sufficiency of greedy sampling in RLHF.