大语言模型的指令微调：一项综述
Instruction Tuning for Large Language Models: A Survey
AI总结：
该综述系统梳理了大语言模型指令微调（即监督微调SFT）的相关研究，涵盖方法、数据集构建、模型训练、多领域应用及影响因素，同时分析了现有缺陷并提出未来研究方向。
摘要：
本文对指令微调（instruction tuning, IT）这一快速发展领域的研究工作展开综述，该技术也可称为监督微调（supervised fine-tuning, SFT）（除非另有说明，本文中监督微调（SFT）与指令微调（IT）可互换使用），是提升大语言模型（large language models, LLMs）能力与可控性的关键技术。指令微调指的是在由（指令，输出）对构成的数据集上以监督方式进一步训练大语言模型的过程，它弥合了大语言模型的下一词预测目标与用户希望大语言模型遵循人类指令的目标之间的差距。在本研究中，我们对相关文献进行了系统梳理，涵盖SFT的通用方法、SFT数据集的构建、SFT模型的训练，以及其在不同模态、领域和应用场景中的应用，同时分析了影响SFT效果的各类因素（例如指令输出的生成、指令数据集的规模等）。我们还回顾了SFT潜在的缺陷及相关批评，以及指出现有策略当前不足的研究工作，并提出了一些有价值的研究方向。项目主页：github.com/xiaoya-li/Instruction-Tuning-Survey
英文摘要：
This paper surveys research works in the quickly advancing field of instruction tuning (IT), which can also be referred to as supervised fine-tuning (SFT)\footnote{In this paper, unless specified otherwise, supervised fine-tuning (SFT) and instruction tuning (IT) are used interchangeably.}, a crucial technique to enhance the capabilities and controllability of large language models (LLMs). Instruction tuning refers to the process of further training LLMs on a dataset consisting of \textsc{(instruction, output)} pairs in a supervised fashion, which bridges the gap between the next-word prediction objective of LLMs and the users' objective of having LLMs adhere to human instructions. In this work, we make a systematic review of the literature, including the general methodology of SFT, the construction of SFT datasets, the training of SFT models, and applications to different modalities, domains and application, along with analysis on aspects that influence the outcome of SFT (e.g., generation of instruction outputs, size of the instruction dataset, etc). We also review the potential pitfalls of SFT along with criticism against it, along with efforts pointing out current deficiencies of existing strategies and suggest some avenues for fruitful research. Project Page: github.com/xiaoya-li/Instruction-Tuning-Survey
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。
