Deliberate then Generate: Enhanced Prompting Framework for Text Generation

TL;DR

The paper introduces the Deliberate then Generate (DTG) prompting framework that enhances large language model performance across diverse text generation tasks by incorporating error detection and candidate correction strategies.

Contribution

It proposes a novel DTG prompting method that improves LLM output quality and achieves state-of-the-art results across multiple text generation tasks with minimal modifications.

Findings

DTG outperforms existing prompting methods consistently.

Achieves state-of-the-art results on multiple datasets.

Effective across diverse text generation tasks.

Abstract

Large language models (LLMs) have shown remarkable success across a wide range of natural language generation tasks, where proper prompt designs make great impacts. While existing prompting methods are normally restricted to providing correct information, in this paper, we encourage the model to deliberate by proposing a novel Deliberate then Generate (DTG) prompting framework, which consists of error detection instructions and candidates that may contain errors. DTG is a simple yet effective technique that can be applied to various text generation tasks with minimal modifications. We conduct extensive experiments on 20+ datasets across 7 text generation tasks, including summarization, translation, dialogue, and more. We show that DTG consistently outperforms existing prompting methods and achieves state-of-the-art performance on multiple text generation tasks. We also provide in-depth analyses to reveal the underlying mechanisms of DTG, which may inspire future research on prompting for LLMs.