Flow-DPO: Improving LLM Mathematical Reasoning through Online Multi-Agent Learning
This addresses the problem of enhancing mathematical reasoning capabilities in LLMs for applications requiring precise and structured outputs, representing an incremental improvement through a novel training approach.
The paper tackles the challenge of generating detailed and accurate reasoning traces for LLMs in mathematical reasoning by introducing Flow-DPO, an online multi-agent learning method that uses incremental output production and DPO with rollouts, resulting in improved performance compared to direct model inference.
Mathematical reasoning is a crucial capability for Large Language Models (LLMs), yet generating detailed and accurate reasoning traces remains a significant challenge. This paper introduces a novel approach to produce high-quality reasoning traces for LLM fine-tuning using online learning \textbf{Flows}. Our method employs an incremental output production Flow, where component LLMs collaboratively construct solutions through iterative communication. We train the Flow using online Direct Preference Optimization (DPO) learning with rollouts, generating DPO pairs for each training example and updating models in real-time. We directly compare the quality of reasoning traces generated by our method with those produced through direct model inference, demonstrating the effectiveness of our approach in improving LLM performance in mathematical reasoning tasks.