New method refines LLM credit assignment for math reasoning

Researchers have developed a new method called Outcome-Grounded Advantage Reshaping (OAR) to improve how large language models handle mathematical reasoning. This technique refines the credit assignment process in reinforcement learning, ensuring that individual reasoning steps are weighted according to their actual impact on the final answer. OAR offers two strategies: one using counterfactual perturbations for high accuracy and another using input-gradient sensitivity for computational efficiency, both significantly outperforming existing methods. AI

影响 Enhances LLM capabilities in complex mathematical reasoning by improving how models learn from their outputs.

排序理由 The cluster contains a research paper detailing a new method for improving LLM reasoning. [lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 阅读 →

• Group Relative Policy Optimization
• Outcome-Grounded Advantage Reshaping
• Ziheng Li

• 论文
• 模型发布

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →