对比指令微调
Contrastive Instruction Tuning
AI总结：
本文提出对比指令微调CoIN，通过拉近语义等价指令对的表示并拉远不同语义指令对的表示，增强大语言模型对未见指令变化的鲁棒性，在PromptBench上平均提升2.5%准确率。
摘要：
指令微调已被视为提升大语言模型（LLMs）在未见任务上性能的一种有前景的方法。然而，当前的大语言模型对未见指令表现出有限的鲁棒性，当同一指令以略有变化的形式或语言风格表述时，会产生不一致的输出。这一行为表明大语言模型缺乏对文本变化的鲁棒性以及对未见指令的泛化能力，可能引发可信性问题。为此，我们提出了对比指令微调（Contrastive Instruction Tuning），该方法最大化语义等价的指令-实例对的隐藏表示之间的相似性，同时最小化语义不同的指令-实例对之间的相似性。为促进这一方法，我们通过对任务指令进行改写来扩充现有的FLAN集合。在PromptBench基准上的实验表明，CoIN在字符、单词、句子和语义层面的变化下，持续提升了大语言模型对未见指令的鲁棒性，准确率平均提高了+2.5%。代码可在https://github.com/luka-group/CoIN获取。
英文摘要：
Instruction tuning has been used as a promising approach to improve the performance of large language models (LLMs) on unseen tasks. However, current LLMs exhibit limited robustness to unseen instructions, generating inconsistent outputs when the same instruction is phrased with slightly varied forms or language styles. This behavior indicates LLMs' lack of robustness to textual variations and generalizability to unseen instructions, potentially leading to trustworthiness issues. Accordingly, we propose Contrastive Instruction Tuning, which maximizes the similarity between the hidden representations of semantically equivalent instruction-instance pairs while minimizing the similarity between semantically different ones. To facilitate this approach, we augment the existing FLAN collection by paraphrasing task instructions. Experiments on the PromptBench benchmark show that CoIN consistently improves LLMs' robustness to unseen instructions with variations across character, word, sentence, and semantic levels by an average of +2.5% in accuracy. Code is available at https://github.com/luka-group/CoIN.
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。
