Compass-Thinker-7B Technical Report
By: Anxiang Zeng , Haibo Zhang , Kaixiang Mo and more
Potential Business Impact:
Teaches computers to solve math problems better.
Recent R1-Zero-like research further demonstrates that reasoning extension has given large language models (LLMs) unprecedented reasoning capabilities, and Reinforcement Learning is the core technology to elicit its complex reasoning. However, conducting RL experiments directly on hyperscale models involves high computational costs and resource demands, posing significant risks. We propose the Compass-Thinker-7B model, which aims to explore the potential of Reinforcement Learning with less computational resources and costs, and provides insights for further research into RL recipes for larger models. Compass-Thinker-7B is trained from an open source model through a specially designed Reinforcement Learning Pipeline. We curate a dataset of 30k verifiable mathematics problems for the Reinforcement Learning Pipeline. By configuring data and training settings with different difficulty distributions for different stages, the potential of the model is gradually released and the training efficiency is improved. Extensive evaluations show that Compass-Thinker-7B possesses exceptional reasoning potential, and achieves superior performance on mathematics compared to the same-sized RL model. Especially in the challenging AIME2024 evaluation, Compass-Thinker-7B achieves 40% accuracy.
Similar Papers
Compass-Thinker-7B Technical Report
Artificial Intelligence
Teaches computers to solve hard math problems.
Generative Adversarial Reasoner: Enhancing LLM Reasoning with Adversarial Reinforcement Learning
Artificial Intelligence
Teaches computers to solve math problems correctly.
A Survey of Reinforcement Learning for Large Reasoning Models
Computation and Language
Teaches computers to think and solve hard problems.