# PSPO*: An Effective Process-supervised Policy Optimization for Reasoning Alignment

arXiv: https://arxiv.org/abs/2411.11681 · HTML: https://mlsanity.com/paper/2411.11681/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning.
- Builds on 5 of its 34 references that are in the map.
- First posted on arXiv 18 Nov 2024.

## Placed in

- LLM reasoning / [Reinforcement learning for reasoning](https://mlsanity.com/lr/reinforcement-learning-for-reasoning/)
- LLM alignment / RLHF / preference optimisation / [Process reward models for reasoning](https://mlsanity.com/alignment/process-reward-models-for-reasoning/)

## Builds on (most cited first, 5 of 5)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [Qwen2 Technical Report](https://mlsanity.com/paper/2407.10671/) (2024)
- [Solving math word problems with process- and outcome-based feedback](https://mlsanity.com/paper/2211.14275/) (2022)

## Built on by (most cited first, 8 of 8)

- [CondAmbigQA: A Benchmark and Dataset for Conditional Ambiguous Question Answering](https://mlsanity.com/paper/2502.01523/) (2025)
- [SoLoPO: Unlocking Long-Context Capabilities in LLMs via Short-to-Long Preference Optimization](https://mlsanity.com/paper/2505.11166/) (2025)
- [Good Learners Think Their Thinking: Generative PRM Makes Large Reasoning Model More Efficient Math Learner](https://mlsanity.com/paper/2507.23317/) (2025)
- [Understanding Reinforcement Learning for Model Training, and future directions with GRAPE](https://mlsanity.com/paper/2509.04501/) (2025)
- [P2S: Probabilistic Process Supervision for General-Domain Reasoning Question Answering](https://mlsanity.com/paper/2601.20649/) (2026)
- [Reinforcing Structured Chain-of-Thought for Video Understanding](https://mlsanity.com/paper/2603.25942/) (2026)
- [PGPO: Potential-Guided Policy Optimization for Multi-Turn Agentic Tasks](https://mlsanity.com/paper/2609.02236/) (2026)
- [Who Gets the Reward & Who Gets the Blame? Evaluation-Aligned Training Signals for Multi-LLM Agents](https://mlsanity.com/paper/2511.10687/) (2025)

## Abstract (arXiv)

Process supervision enhances the performance of large language models in reasoning tasks by providing feedback at each step of chain-of-thought reasoning. However, due to the lack of effective process supervision methods, even advanced large language models are prone to logical errors and redundant reasoning. We claim that the effectiveness of process supervision significantly depends on both the accuracy and the length of reasoning chains. Moreover, we identify that these factors exhibit a nonlinear relationship with the overall reward score of the reasoning process. Inspired by these insights, we propose a novel process supervision paradigm, PSPO*, which systematically outlines the workflow from reward model training to policy optimization, and highlights the importance of nonlinear rewards in process supervision. Based on PSPO*, we develop the PSPO-WRS, which considers the number of reasoning steps in determining reward scores and utilizes an adjusted Weibull distribution for nonlinear reward shaping. Experimental results on six mathematical reasoning datasets demonstrate that PSPO-WRS consistently outperforms current mainstream models.