Flatness-Aware Prompt Selection Improves Accuracy and Sample Efficiency

Lingfeng Shen, Weiting Tan, Boyuan Zheng, Daniel Khashabi

Submission Type: Regular Long Paper

Submission Track: Language Modeling and Analysis of Language Models

Submission Track 2: Machine Learning for NLP

Keywords: prompt selection, flatness of prompt

Abstract:

With growing capabilities of large language models, prompting them has become the dominant way to access them.
This has motivated the development of strategies for automatically selecting effective language prompts.
In this paper, we introduce **pFlat** (prompt flatness), a new metric to quantify the expected utility of a language prompt. This metric is
inspired by *flatness* regularization in statistical learning that quantifies the robustness of the model towards its parameter perturbations. We provide theoretical foundations for this metric and its relationship with other prompt selection metrics, providing a comprehensive understanding of existing methods.
Empirically, we show that
combining **pFlat** with existing metrics
improves
both performance and sample efficiency.
Our metric outperforms the previous prompt selection metrics with an average increase of 10% in Pearson correlation across 6 classification benchmarks, and the prompt selected by our metric gains 5% higher accuracy than previous metrics across the benchmarks.

Submission Number: 2124