Automatic Prompt Engineering with No Task Cues and No Tuning

TL;DR

This paper introduces a simple, tuning-free automatic prompt engineering system that effectively handles cryptic column name expansion in multilingual database tables, marking a novel application in non-English contexts.

Contribution

The work presents the first application of automatic prompt engineering to cryptic column name expansion and to languages other than English, with a straightforward design and no task-specific tuning.

Findings

Effective on cryptic column name expansion in English and German

No tuning or explicit task clues needed for high performance

First to apply automatic prompt engineering to this task and language set

Abstract

This paper presents a system for automatic prompt engineering that is much simpler in both design and application and yet as effective as the existing approaches. It requires no tuning and no explicit clues about the task. We evaluated our approach on cryptic column name expansion (CNE) in database tables, a task which is critical for tabular data search, access, and understanding and yet there has been very little existing work. We evaluated on datasets in two languages, English and German. This is the first work to report on the application of automatic prompt engineering for the CNE task. To the best of our knowledge, this is also the first work on the application of automatic prompt engineering for a language other than English.