Evoke: Evoking Critical Thinking Abilities in LLMs via Reviewer-Author Prompt Editing

TL;DR

Evoke is an automatic prompt refinement framework that uses a dual-LLM reviewer-author loop and hard sample selection to enhance LLM performance, notably improving logical fallacy detection accuracy.

Contribution

The paper introduces Evoke, a novel prompt refinement method employing a reviewer-author feedback loop and hard sample focus, significantly boosting LLM task performance.

Findings

Evoke outperforms existing prompt methods in logical fallacy detection.

Evoke achieves above 80 accuracy, compared to below 20 for baselines.

The feedback loop effectively refines prompts for better task understanding.

Abstract

Large language models (LLMs) have made impressive progress in natural language processing. These models rely on proper human instructions (or prompts) to generate suitable responses. However, the potential of LLMs are not fully harnessed by commonly-used prompting methods: many human-in-the-loop algorithms employ ad-hoc procedures for prompt selection; while auto prompt generation approaches are essentially searching all possible prompts randomly and inefficiently. We propose Evoke, an automatic prompt refinement framework. In Evoke, there are two instances of a same LLM: one as a reviewer (LLM-Reviewer), it scores the current prompt; the other as an author (LLM-Author), it edits the prompt by considering the edit history and the reviewer's feedback. Such an author-reviewer feedback loop ensures that the prompt is refined in each iteration. We further aggregate a data selection approach to Evoke, where only the hard samples are exposed to the LLM. The hard samples are more important because the LLM can develop deeper understanding of the tasks out of them, while the model may already know how to solve the easier cases. Experimental results show that Evoke significantly outperforms existing methods. For instance, in the challenging task of logical fallacy detection, Evoke scores above 80, while all other baseline methods struggle to reach 20.