基于人类课程的指令微调
Instruction Tuning with Human Curriculum
AI总结：
本文提出课程指令微调及模拟人类教育进程的合成指令数据生成框架，证明仅按课程顺序组织数据即可在九个基准上稳定提升性能且无额外计算成本。
摘要：
在本工作中，我们（1）提出了课程指令微调（Curriculum Instruction Tuning），（2）探索了采用多样化课程策略的潜在优势，并且（3）描述了一个合成指令-响应生成框架，以补充我们的理论方法。与现有指令微调数据集不同，我们的生成流程经过系统化构建，用以模拟人类学习循序渐进、井然有序的特征。此外，我们还描述了一种生成指令-响应数据集的方法，该方法利用教育学科目录，广泛覆盖从中学到研究生阶段的人类教育各个阶段。在训练之前，我们精心组织指令数据，以确保问题在（A）学科内容和（B）指令复杂度两方面的难度逐步提升。研究结果表明，与随机打乱相比，仅对指令数据应用课程排序即可实现显著的性能提升（在 TruthfulQA 上提高 +4.76，在 MMLU 上提高 +2.98，在 OpenbookQA 上提高 +2.8，在 ARC-hard 上提高 +1.28）。这种提升不会产生额外的计算开销。通过全面实验，我们观察到所提方法的优势在九个基准测试中持续显现。
英文摘要：
In this work, we (1) introduce Curriculum Instruction Tuning, (2) explore the potential advantages of employing diverse curriculum strategies, and (3) delineate a synthetic instruction-response generation framework that complements our theoretical approach. Distinct from the existing instruction tuning dataset, our generation pipeline is systematically structured to emulate the sequential and orderly characteristic of human learning. Additionally, we describe a methodology for generating instruction-response datasets that extensively span the various stages of human education, from middle school through the graduate level, utilizing educational subject catalogs. Before training, we meticulously organize the instruction data to ensure that questions escalate in difficulty regarding (A) the subject matter and (B) the intricacy of the instructions. The findings of our study reveal that substantial improvements in performance can be achieved through the mere application of curriculum ordering to instruction data (achieving gains of +4.76 on TruthfulQA, +2.98 on MMLU, +2.8 on OpenbookQA, and +1.28 on ARC-hard) compared to random shuffling. This enhancement is achieved without incurring additional computational expenses. Through comprehensive experimentation, we observe that the advantages of our proposed method are consistently evident across nine benchmarks.
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。
