# RS-DPO: A Hybrid Rejection Sampling and Direct Preference Optimization Method for Alignment of Large Language Models

arXiv: https://arxiv.org/abs/2402.10038 · HTML: https://mlsanity.com/paper/2402.10038/

## Key facts

- Cited by 48 papers in the map (data through 14 Oct 2026).
- Placed in LLM preference optimization methods, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 25 of its 34 references that are in the map.
- First posted on arXiv 15 Feb 2024.

## Placed in

- LLM alignment / RLHF / preference optimisation / [LLM preference optimization methods](https://mlsanity.com/alignment/llm-preference-optimization-methods/)

## Builds on (most cited first, 25 of 25)

- [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)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [Scaling Instruction-Finetuned Language Models](https://mlsanity.com/paper/2210.11416/) (2022)
- [Self-Instruct: Aligning Language Models with Self-Generated Instructions](https://mlsanity.com/paper/2212.10560/) (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)
- [Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling](https://mlsanity.com/paper/2304.01373/) (2023)
- [WebGPT: Browser-assisted question-answering with human feedback](https://mlsanity.com/paper/2112.09332/) (2021)
- [OpenAssistant Conversations -- Democratizing Large Language Model Alignment](https://mlsanity.com/paper/2304.07327/) (2023)
- [RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment](https://mlsanity.com/paper/2304.06767/) (2023)
- [RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback](https://mlsanity.com/paper/2309.00267/) (2023)
- [Reinforced Self-Training (ReST) for Language Modeling](https://mlsanity.com/paper/2308.08998/) (2023)
- [Statistical Rejection Sampling Improves Preference Optimization](https://mlsanity.com/paper/2309.06657/) (2023)
- [Understanding Dataset Difficulty with V-Usable Information](https://mlsanity.com/paper/2110.08420/) (2021)
- [A Long Way to Go: Investigating Length Correlations in RLHF](https://mlsanity.com/paper/2310.03716/) (2023)
- [Efficient RLHF: Reducing the Memory Usage of PPO](https://mlsanity.com/paper/2309.00754/) (2023)

## Built on by (most cited first, 30 of 48)

- [Preference Fine-Tuning of LLMs Should Leverage Suboptimal, On-Policy Data](https://mlsanity.com/paper/2404.14367/) (2024)
- [Is Best-of-N the Best of Them? Coverage, Scaling, and Optimality in Inference-Time Alignment](https://mlsanity.com/paper/2503.21878/) (2025)
- [PersonaMem-v2: Towards Personalized Intelligence via Learning Implicit User Personas and Agentic Memory](https://mlsanity.com/paper/2512.06688/) (2025)
- [Sample More to Think Less: Group Filtered Policy Optimization for Concise Reasoning](https://mlsanity.com/paper/2508.09726/) (2025)
- [TreeBoN: Enhancing Inference-Time Alignment with Speculative Tree-Search and Best-of-N Sampling](https://mlsanity.com/paper/2410.16033/) (2024)
- [Cascade Reward Sampling for Efficient Decoding-Time Alignment](https://mlsanity.com/paper/2406.16306/) (2024)
- [A Survey of Direct Preference Optimization](https://mlsanity.com/paper/2503.11701/) (2025)
- [LongReward: Improving Long-context Large Language Models with AI Feedback](https://mlsanity.com/paper/2410.21252/) (2024)
- [The Crucial Role of Samplers in Online Direct Preference Optimization](https://mlsanity.com/paper/2409.19605/) (2024)
- [A Comprehensive Survey of Direct Preference Optimization: Datasets, Theories, Variants, and Applications](https://mlsanity.com/paper/2410.15595/) (2024)
- [CasaGPT: Cuboid Arrangement and Scene Assembly for Interior Design](https://mlsanity.com/paper/2504.19478/) (2025)
- [Token-Importance Guided Direct Preference Optimization](https://mlsanity.com/paper/2505.19653/) (2025)
- [Right Now, Wrong Then: Non-Stationary Direct Preference Optimization under Preference Drift](https://mlsanity.com/paper/2407.18676/) (2024)
- [R.I.P.: Better Models by Survival of the Fittest Prompts](https://mlsanity.com/paper/2501.18578/) (2025)
- [Larger or Smaller Reward Margins to Select Preferences for Alignment?](https://mlsanity.com/paper/2503.01864/) (2025)
- [Understanding the Performance Gap in Preference Learning: A Dichotomy of RLHF and DPO](https://mlsanity.com/paper/2505.19770/) (2025)
- [AIR: A Systematic Analysis of Annotations, Instructions, and Response Pairs in Preference Dataset](https://mlsanity.com/paper/2504.03612/) (2025)
- [Intuitive Fine-Tuning: Towards Simplifying Alignment into a Single Process](https://mlsanity.com/paper/2405.11870/) (2024)
- [Systematic Reward Gap Optimization for Mitigating VLM Hallucinations](https://mlsanity.com/paper/2411.17265/) (2024)
- [OrthAlign: Orthogonal Subspace Decomposition for Non-Interfering Multi-Objective Alignment](https://mlsanity.com/paper/2509.24610/) (2025)
- [TeaRAG: A Token-Efficient Agentic Retrieval-Augmented Generation Framework](https://mlsanity.com/paper/2511.05385/) (2025)
- [Reward Modeling for Reinforcement Learning-Based LLM Reasoning: Design, Challenges, and Evaluation](https://mlsanity.com/paper/2602.09305/) (2026)
- [Sparse Rewards Can Self-Train Dialogue Agents](https://mlsanity.com/paper/2409.04617/) (2024)
- [Refining Alignment Framework for Diffusion Models with Intermediate-Step Preference Ranking](https://mlsanity.com/paper/2502.01667/) (2025)
- [Learning a Pessimistic Reward Model in RLHF](https://mlsanity.com/paper/2505.20556/) (2025)
- [Scaling-up Perceptual Video Quality Assessment](https://mlsanity.com/paper/2505.22543/) (2025)
- [Unified Preference Optimization: Language Model Alignment Beyond the Preference Frontier](https://mlsanity.com/paper/2405.17956/) (2024)
- [Efficient Safety Retrofitting Against Jailbreaking for LLMs](https://mlsanity.com/paper/2502.13603/) (2025)
- [Less is More: Improving LLM Alignment via Preference Data Selection](https://mlsanity.com/paper/2502.14560/) (2025)
- [RePO: Understanding Preference Learning Through ReLU-Based Optimization](https://mlsanity.com/paper/2503.07426/) (2025)

## Abstract (arXiv)

Reinforcement learning from human feedback (RLHF) has been extensively employed to align large language models with user intent. However, proximal policy optimization (PPO) based RLHF is occasionally unstable requiring significant hyperparameter finetuning, and computationally expensive to maximize the estimated reward during alignment. Recently, direct preference optimization (DPO) is proposed to address those challenges. However, DPO relies on contrastive responses generated from human annotator and alternative LLM, instead of the policy model, limiting the effectiveness of the RLHF. In this paper, we addresses both challenges by systematically combining rejection sampling (RS) and DPO. Our proposed method, RS-DPO, initiates with the development of a supervised fine-tuned policy model (SFT). A varied set of k responses per prompt are sampled directly from the SFT model. RS-DPO identifies pairs of contrastive samples based on their reward distribution. Finally, we apply DPO with the contrastive samples to align the model to human preference. Our experiments indicate that our proposed method effectively fine-tunes LLMs with limited resource environments, leading to improved alignment with user intent. Furthermore, it outperforms existing methods, including RS, PPO, and DPO.