Skill-Based Few-Shot Selection for In-Context Learning

Shengnan An, Bo Zhou, Zeqi Lin, Qiang Fu, Bei Chen, Nanning Zheng, Weizhu Chen, Jian-Guang Lou

Submission Type: Regular Long Paper

Submission Track: Theme Track: Large Language Models and the Future of NLP

Submission Track 2: Syntax, Parsing and their Applications

Keywords: In-Context Learning, Few-Shot Selection, Semantic Parsing

TL;DR: We propose Skill-KNN, a skill-based few-shot selection method for in-context learning, to facilitate semantic parsing. Skill-KNN significantly outperforms raw-input-based selection methods on semantic parsing tasks.

Abstract:

*In-context learning* is the paradigm that adapts large language models to downstream tasks by providing a few examples.
*Few-shot selection*---selecting appropriate examples for each test instance separately---is important for in-context learning.
In this paper, we propose **Skill-KNN**, a skill-based few-shot selection method for in-context learning.
The key advantages of Skill-KNN include: (1) it addresses the problem that existing methods based on pre-trained embeddings can be easily biased by surface natural language features that are not important for the target task; (2) it does not require training or fine-tuning of any models, making it suitable for frequently expanding or changing example banks.
The key insight is to optimize the inputs fed into the embedding model, rather than tuning the model itself.
Technically, Skill-KNN generates the skill-based descriptions for each test case and candidate example by utilizing a pre-processing few-shot prompting, thus eliminating unimportant surface features.
Experimental results across five cross-domain semantic parsing datasets and six backbone models show that Skill-KNN significantly outperforms existing methods.

Submission Number: 460