PACIT：释放示例的力量以实现更好的上下文指令微调
PACIT: Unlocking the Power of Examples for Better In-Context Instruction Tuning
AI总结：
受“期望难度”启发，提出 PACIT 上下文指令微调方法，通过让模型主动验证正负例的正确性作为生成响应的条件，显著提升了大语言模型在域内外任务上的性能。
摘要：
指令微调通过使用有监督指令数据进行微调，增强了大型语言模型遵循指令的能力。先前的工作提出了上下文指令微调（ICIT），其中将特定的正例或负例纳入提示中以获得更好的性能。在这项工作中，受“期望难度”这一教学概念的启发，我们提出了一种简单有效的上下文指令微调方法 PACIT。PACIT 方法通过鼓励模型主动学习掌握正例和负例之间的区别，而不是仅仅阅读，从而释放了示例的力量。模型被期望首先根据任务描述验证所提供示例的正确性，然后将其作为为任务实例生成更好响应的条件。我们广泛的实验证明了 PACIT 的有效性，在域内和域外任务上分别以高达 9.16 和 3.14 的平均 ROUGE-L 分数优于 ICIT 基线。此外，即使所有正例和负例都是使用 self-instruct 方法生成的，PACIT 也能显著增强指令微调的性能。
英文摘要：
Instruction tuning enhances the instruction following ability of large language models by finetuning with supervised instruction data. Previous work proposes in-context instruction tuning (ICIT) where specific positive or negative examples are incorporated into the prompt for better performance. In this work, we propose PACIT, a simple and effective in-context instruction tuning method, inspired by the pedagogical concept of desirable difficulty. The PACIT method unlocks the power of examples by encouraging the model to actively learn to grasp the distinctions between the positive and negative examples instead of merely reading. The model is expected to first verify the correctness of the provided example according to the task description, which is then set as the condition for generating a better response to the task instance. Our extensive experiments prove the effectiveness of PACIT, outperforming ICIT baseline on both in-domain and out-domain tasks up to 9.16 and 3.14 average ROUGE-L scores, respectively. Moreover, PACIT can notably enhance the performance of instruction tuning even when all positive and negative examples are generated with a self-instruct method.
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。
