arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2310.07488cs.CLcs.AIcs.LG
收藏

KwaiYiiMath:技术报告

KwaiYiiMath: Technical Report

Kuaishou Technology(快手科技)
Jiayi Fu, Lei Lin, Xiaoyang Gao, Pengli Liu, Zhengzong Chen, Zhirui Yang, Shengnan Zhang, Xue Zheng, Yan Li, Yuliang Liu, Xucheng Ye, Yiqiao Liao, Chao Liao, Bi… 展开作者

Jiayi Fu, Lei Lin, Xiaoyang Gao, Pengli Liu, Zhengzong Chen, Zhirui Yang, Shengnan Zhang, Xue Zheng, Yan Li, Yuliang Liu, Xucheng Ye, Yiqiao Liao, Chao Liao, Bin Chen, Chengru Song, Junchen Wan, Zijia Lin, Fuzheng Zhang, Zhongyuan Wang, Di Zhang, Kun Gai

AI总结:

本报告提出KwaiYiiMath,通过监督微调和人类反馈强化学习增强KwaiYiiBase1的数学推理能力,并构建KMath测试集,在多个基准上取得同类规模模型的最优性能。

摘要:

大型语言模型(LLMs)的最新进展展示了其在处理各种自然语言处理(NLP)下游任务中的卓越能力,甚至在需要多步推理的数学任务上也表现不俗。在本报告中,我们介绍了KwaiYiiMath,它通过应用监督微调(SFT)和人类反馈强化学习(RLHF),增强了KwaiYiiBase1的数学推理能力,涵盖了英文和中文数学任务。同时,我们还构建了一个小规模的中文小学数学测试集(名为KMath),包含188个示例,用于评估模型生成的解题过程的正确性。实证研究表明,与相似规模的模型相比,KwaiYiiMath在GSM8k、CMath和KMath上分别达到了最先进(SOTA)的性能。

英文摘要:

Recent advancements in large language models (LLMs) have demonstrated remarkable abilities in handling a variety of natural language processing (NLP) downstream tasks, even on mathematical tasks requiring multi-step reasoning. In this report, we introduce the KwaiYiiMath which enhances the mathematical reasoning abilities of KwaiYiiBase1, by applying Supervised Fine-Tuning (SFT) and Reinforced Learning from Human Feedback (RLHF), including on both English and Chinese mathematical tasks. Meanwhile, we also constructed a small-scale Chinese primary school mathematics test set (named KMath), consisting of 188 examples to evaluate the correctness of the problem-solving process generated by the models. Empirical studies demonstrate that KwaiYiiMath can achieve state-of-the-art (SOTA) performance on GSM8k, CMath, and KMath compared with the similar size models, respectively.

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。

↑