MAIN：互斥对齐是指令微调的必要条件
MAIN: Mutual Alignment Is Necessary for instruction tuning
AI总结：
本文提出MAIN框架，通过互斥约束增强指令与响应的一致性，提升LLM在多种基准上的性能。
摘要：
指令微调已使大型语言模型（LLMs）实现了显著的性能提升，但其成功高度依赖于大量高质量的指令-响应对的可用性。为满足这一需求，各种方法已被开发出来以大规模合成数据。然而，当前用于扩大数据生成的方法往往忽视了一个关键方面：指令与响应之间的对齐。我们假设指令-响应对的质量并非由每个组件的单独质量决定，而是由指令与响应之间的互斥对齐程度决定。为此，我们提出了一个互斥对齐框架（MAIN），通过互斥约束来强制指令与响应之间的一致性。我们证明MAIN在不同的模型架构和大小上具有良好的泛化能力，在LLaMA、Mistral和Qwen模型上实现了跨多样基准的最先进性能。本工作强调了指令-响应对齐在使LLMs实现可推广且高质量的指令微调中的关键作用。所有代码均可从我们的仓库中获得。
英文摘要：
Instruction tuning has empowered large language models (LLMs) to achieve remarkable performance, yet its success heavily depends on the availability of large-scale, high-quality instruction-response pairs. To meet this demand, various methods have been developed to synthesize data at scale. However, current methods for scaling up data generation often overlook a crucial aspect: the alignment between instructions and responses. We hypothesize that the quality of instruction-response pairs is determined not by the individual quality of each component, but by the degree of mutual alignment. To address this, we propose a Mutual Alignment Framework (MAIN) which enforces coherence between instructions and responses through mutual constraints. We demonstrate that MAIN generalizes well across model architectures and sizes, achieving state-of-the-art performance on LLaMA, Mistral, and Qwen models across diverse benchmarks. This work underscores the critical role of instruction-response alignment in enabling generalizable and high-quality instruction tuning for LLMs. All code is available from our repository.
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。
