Value-Guided Search for Efficient Chain-of-Thought Reasoning
Kaiwen Wang, Jin Peng Zhou, Jonathan Daniel Chang, Zhaolin Gao, Nathan Kallus, Kianté Brantley, Wen Sun
Abstract
In this paper, we propose a simple and efficient method for value model training on long-context reasoning traces. Compared to existing process reward models (PRMs), our method does not require a fine-grained notion of ``step,'' which is difficult to define for long-context reasoning models. By collecting a dataset of 2.5 million reasoning traces, we train a 1.5B token-level value model and apply it to DeepSeek models for improved performance with test-time compute scaling. We find that block-wise value-guided search (\texttt{VGS}) with a final weighted majority vote achieves better test-time scaling than standard methods such as majority voting or best-of-$n$. Moreover, \texttt{VGS} significantly reduces the inference FLOPs required to achieve the same performance of majority voting. Our dataset, model and codebase are open-sourced at \codeurl.
BibTeX
@inproceedings{
wang2025valueguided,
title={Value-Guided Search for Efficient Chain-of-Thought Reasoning},
author={Kaiwen Wang and Jin Peng Zhou and Jonathan Daniel Chang and Zhaolin Gao and Nathan Kallus and Kiant{\'e} Brantley and Wen Sun},
booktitle={The Thirty-ninth Annual Conference on Neural Information Processing Systems},
year={2025},
url={https://openreview.net/forum?id=jOsuKwiCL0}
}