# SGPO: Self-Generated Preference Optimization based on Self-Improver

arXiv: https://arxiv.org/abs/2507.20181 · HTML: https://mlsanity.com/paper/2507.20181/

## Key facts

- Cited by 1 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 20 of its 37 references that are in the map.
- First posted on arXiv 27 Jul 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [LLM preference optimization methods](https://mlsanity.com/alignment/llm-preference-optimization-methods/)
- Synthetic data and self-improvement for LLMs / [Self-rewarding alignment and self-training](https://mlsanity.com/synth/self-rewarding-alignment-and-self-training/)

## Builds on (most cited first, 20 of 20)

- [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)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Qwen2 Technical Report](https://mlsanity.com/paper/2407.10671/) (2024)
- [SimPO: Simple Preference Optimization with a Reference-Free Reward](https://mlsanity.com/paper/2405.14734/) (2024)
- [Enhancing Chat Language Models by Scaling High-quality Instructional Conversations](https://mlsanity.com/paper/2305.14233/) (2023)
- [ORPO: Monolithic Preference Optimization without Reference Model](https://mlsanity.com/paper/2403.07691/) (2024)
- [Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models](https://mlsanity.com/paper/2401.01335/) (2024)
- [From Crowdsourced Data to High-Quality Benchmarks: Arena-Hard and BenchBuilder Pipeline](https://mlsanity.com/paper/2406.11939/) (2024)
- [KTO: Model Alignment as Prospect Theoretic Optimization](https://mlsanity.com/paper/2402.01306/) (2024)
- [Meta-Rewarding Language Models: Self-Improving Alignment with LLM-as-a-Meta-Judge](https://mlsanity.com/paper/2407.19594/) (2024)
- [Qwen2.5-1M Technical Report](https://mlsanity.com/paper/2501.15383/) (2025)
- [β-DPO: Direct Preference Optimization with Dynamic β](https://mlsanity.com/paper/2407.08639/) (2024)
- [TLCR: Token-Level Continuous Reward for Fine-grained Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2407.16574/) (2024)
- [Self-Boosting Large Language Models with Synthetic Preference Data](https://mlsanity.com/paper/2410.06961/) (2024)
- [Policy Optimization in RLHF: The Impact of Out-of-preference Data](https://mlsanity.com/paper/2312.10584/) (2023)
- [Aligning Large Language Models by On-Policy Self-Judgment](https://mlsanity.com/paper/2402.11253/) (2024)

## Built on by (most cited first, 1 of 1)

- [CRPO: Character-centric Group Relative Policy Optimization for Role-aware Reasoning in Role-playing Agents](https://mlsanity.com/paper/2605.25511/) (2026)

## Abstract (arXiv)

Large language models (LLMs), despite their extensive pretraining on diverse datasets, require effective alignment to human preferences for practical and reliable deployment. Conventional alignment methods typically employ off-policy learning and depend on human-annotated datasets, which limits their broad applicability and introduces distribution shift issues during training. To address these challenges, we propose Self-Generated Preference Optimization based on Self-Improver (SGPO), an innovative alignment framework that leverages an on-policy self-improving mechanism. Specifically, the improver refines responses from a policy model to self-generate preference data for direct preference optimization (DPO) of the policy model. Here, the improver and policy are unified into a single model, and in order to generate higher-quality preference data, this self-improver learns to make incremental yet discernible improvements to the current responses by referencing supervised fine-tuning outputs. Experimental results on AlpacaEval 2.0 and Arena-Hard show that the proposed SGPO significantly improves performance over DPO and baseline self-improving methods without using external preference data.