# Proxy-RLHF: Decoupling Generation and Alignment in Large Language Model with Proxy

arXiv: https://arxiv.org/abs/2403.04283 · HTML: https://mlsanity.com/paper/2403.04283/

## Key facts

- Cited by 2 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 17 of its 18 references that are in the map.
- First posted on arXiv 7 Mar 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, 17 of 17)

- [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)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [Scaling Instruction-Finetuned Language Models](https://mlsanity.com/paper/2210.11416/) (2022)
- [Learning to summarize from human feedback](https://mlsanity.com/paper/2009.01325/) (2020)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2307.15217/) (2023)
- [VoxPoser: Composable 3D Value Maps for Robotic Manipulation with Language Models](https://mlsanity.com/paper/2307.05973/) (2023)
- [RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment](https://mlsanity.com/paper/2304.06767/) (2023)
- [Safe RLHF: Safe Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2310.12773/) (2023)
- [RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback](https://mlsanity.com/paper/2309.00267/) (2023)
- [LLM+P: Empowering Large Language Models with Optimal Planning Proficiency](https://mlsanity.com/paper/2304.11477/) (2023)
- [Is Reinforcement Learning (Not) for Natural Language Processing: Benchmarks, Baselines, and Building Blocks for Natural Language Policy Optimization](https://mlsanity.com/paper/2210.01241/) (2022)
- [An Empirical Survey on Long Document Summarization: Datasets, Models and Metrics](https://mlsanity.com/paper/2207.00939/) (2022)

## Built on by (most cited first, 2 of 2)

- [Systematic Evaluation of LLM-as-a-Judge in LLM Alignment Tasks: Explainable Metrics and Diverse Prompt Templates](https://mlsanity.com/paper/2408.13006/) (2024)
- [On the Transformations across Reward Model, Parameter Update, and In-Context Prompt](https://mlsanity.com/paper/2406.16377/) (2024)

## Abstract (arXiv)

Reinforcement Learning from Human Feedback (RLHF) is the prevailing approach to ensure Large Language Models (LLMs) align with human values. However, existing RLHF methods require a high computational cost, one main reason being that RLHF assigns both the generation and alignment tasks to the LLM simultaneously. In this paper, we introduce Proxy-RLHF, which decouples the generation and alignment processes of LLMs, achieving alignment with human values at a much lower computational cost. We start with a novel Markov Decision Process (MDP) designed for the alignment process and employ Reinforcement Learning (RL) to train a streamlined proxy model that oversees the token generation of the LLM, without altering the LLM itself. Experiments show that our method achieves a comparable level of alignment with only 1% of the training parameters of other methods.