Toward Human Readable Prompt Tuning: Kubrick’s The Shining is a good movie, and a good prompt too?
Weijia Shi Xiaochuang Han Hila Gonen Ari Holtzman Yulia Tsvetkov Luke Zettlemoyer Paul G. Allen School of Computer Science、Engineering, University of Washington, Seattle, WA {swj0419, xhan77, hilagnn, ahai, yuliats, Equal contribution. Order randomly determined.
Abstract
Large language models can perform new tasks in a zero-shot fashion, given natural language prompts that specify the desired behavior. Such prompts are typically hand engineered, but can also be learned with gradient-based methods from labeled data. However, it is underexplored what factors make the prompts effective, especially when the prompts are natural language. In this paper, we investigate common attributes shared by effective prompts. We first propose a human readable prompt tuning method (FluentPrompt) based on Langevin dynamics that incorporates a fluency constraint to find a diverse distribution of effective and fluent prompts. Our analysis reveals that effective prompts are topically related to the task domain and calibrate the prior probability of label words. Based on these findings, we also propose a method for generating prompts using only unlabeled data, outperforming strong baselines by an average of 7.0% accuracy across three tasks.
中文速览
大语言模型能够依靠提示语（prompt）完成各种任务，但究竟是什么让一条提示语奏效，学界一直缺乏清晰认识。研究者提出了一种基于朗之万动力学（Langevin dynamics）的流畅提示调优方法 FluentPrompt，它在搜索高效提示的同时引入困惑度约束，使生成的提示既性能优良又符合自然语言表达，从而为分析提供了大量可读的样本。通过对这批提示的系统分析，他们发现有效提示有两个共同特征：与任务领域高度相关，以及能够校准（calibrate）模型对标签词的先验概率分布。基于这两条发现，他们进一步提出了无监督版本 Unsupervised FluentPrompt，仅用无标注数据就能自动发现优质提示，在三个任务上平均准确率比强基线高出 7.0%，为在没有标注资源的情况下自动构建有效提示提供了新思路。
原文 arXiv:2212.10539;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2212.10539v1