rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinking
Guan, Xinyu, Zhang, Li Lyna, Liu, Yifei, Shang, Ning, Sun, Youran, Zhu, Yi, Yang, Fan, Yang, Mao
–arXiv.org Artificial Intelligence
We present rStar-Math to demonstrate that small language models (SLMs) can rival or even surpass the math reasoning capability of OpenAI o1, without distillation from superior models. rStar-Math achieves this by exercising "deep thinking" through Monte Carlo Tree Search (MCTS), where a math policy SLM performs test-time search guided by an SLM-based process reward model. rStar-Math introduces three innovations to tackle the challenges in training the two SLMs: (1) a novel code-augmented CoT data sythesis method, which performs extensive MCTS rollouts to generate step-by-step verified reasoning trajectories used to train the policy SLM; (2) a novel process reward model training method that avoids na\"ive step-level score annotation, yielding a more effective process preference model (PPM); (3) a self-evolution recipe in which the policy SLM and PPM are built from scratch and iteratively evolved to improve reasoning capabilities. Through 4 rounds of self-evolution with millions of synthesized solutions for 747k math problems, rStar-Math boosts SLMs' math reasoning to state-of-the-art levels. On the MATH benchmark, it improves Qwen2.5-Math-7B from 58.8% to 90.0% and Phi3-mini-3.8B from 41.4% to 86.4%, surpassing o1-preview by +4.5% and +0.9%. On the USA Math Olympiad (AIME), rStar-Math solves an average of 53.3% (8/15) of problems, ranking among the top 20% the brightest high school math students. Code and data will be available at https://github.com/microsoft/rStar.
arXiv.org Artificial Intelligence
Jan-8-2025
- Country:
- Asia (0.28)
- North America > United States (0.24)
- Genre:
- Research Report
- New Finding (0.46)
- Promising Solution (0.67)
- Research Report
- Industry:
- Education > Curriculum > Subject-Specific Education (1.00)
- Technology: