arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2402.02658cs.AIcs.CLcs.LG
收藏

通过验证器进行多步问题求解:基于模型诱导的过程监督的实证分析

Multi-step Problem Solving Through a Verifier: An Empirical Analysis on Model-induced Process Supervision

University of California, San Diego(加利福尼亚大学圣迭戈分校);Google(谷歌公司)

Zihan Wang, Yunxuan Li, Yuexin Wu, Liangchen Luo, Le Hou, Hongkun Yu, Jingbo Shang

AI总结:

提出自动化数据整理方法MiPS,通过推理模型采样补全标注中间步骤准确率,优先关注验证器高预测分数,显著提升PaLM 2在数学和编程任务上的性能并展现强泛化能力。

摘要:

过程监督通过使用训练好的验证器评估推理器生成的中间步骤,在多步问题求解中展现出显著提升。为避免人工标注验证器训练数据的高昂成本,我们提出了一种自动化数据整理的新方法——模型诱导过程监督(MiPS)。MiPS通过推理模型对某解法的中间步骤采样补全,并获取定义为正确补全比例的准确率来标注该步骤。推理器的不准确会导致MiPS低估中间步骤的准确率,因此我们建议并实证表明,与以往基于人工整理数据的观察相反,应优先关注验证器高预测分数的验证,而非低预测分数的验证。我们的方法显著提升了PaLM 2在数学和编程任务上的性能(与输出监督训练的验证器相比,GSM8K准确率+0.67%,MATH+4.16%,MBPP+0.92%)。此外,研究表明该验证器在不同推理模型间展现出强大的泛化能力。

英文摘要:

Process supervision, using a trained verifier to evaluate the intermediate steps generated by a reasoner, has demonstrated significant improvements in multi-step problem solving. In this paper, to avoid the expensive effort of human annotation on the verifier training data, we introduce Model-induced Process Supervision (MiPS), a novel method for automating data curation. MiPS annotates an intermediate step by sampling completions of this solution through the reasoning model, and obtaining an accuracy defined as the proportion of correct completions. Inaccuracies of the reasoner would cause MiPS underestimating the accuracy of intermediate steps, therefore, we suggest and empirically show that verification focusing on high predicted scores of the verifier shall be preferred over that of low predicted scores, contrary to prior observations on human curated data. Our approach significantly improves the performance of PaLM 2 on math and coding tasks (accuracy +0.67% on GSM8K, +4.16% on MATH, +0.92% on MBPP compared with an output supervision trained verifier). Additionally, our study demonstrates that the verifier exhibits strong generalization ability across different reasoning models.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑