kNN-ICL: Compositional Task-Oriented Parsing Generalization with Nearest Neighbor In-Context Learning

TL;DR

This paper introduces kNN-ICL, a method that enhances large language models' ability to perform semantic parsing in task-oriented dialogue by combining in-context learning with nearest neighbor retrieval, improving understanding of complex commands.

Contribution

The paper proposes kNN-ICL, a novel approach that simplifies prompt design and improves semantic parsing performance by integrating nearest neighbor retrieval with in-context learning.

Findings

Simple ICL achieves performance comparable to supervised models.

kNN-ICL significantly improves understanding of complex requests.

Enhancement occurs without extra data or specialized prompts.

Abstract

Task-Oriented Parsing (TOP) enables conversational assistants to interpret user commands expressed in natural language, transforming them into structured outputs that combine elements of both natural language and intent/slot tags. Recently, Large Language Models (LLMs) have achieved impressive performance in synthesizing computer programs based on a natural language prompt, mitigating the gap between natural language and structured programs. Our paper focuses on harnessing the capabilities of LLMs for semantic parsing tasks, addressing the following three key research questions: 1) How can LLMs be effectively utilized for semantic parsing tasks? 2) What defines an effective prompt? and 3) How can LLM overcome the length constraint and streamline prompt design by including all examples as prompts? We introduce k Nearest Neighbor In-Context Learning(kNN-ICL), which simplifies prompt engineering by allowing it to be built on top of any design strategy while providing access to all demo examples. Extensive experiments show that: 1)Simple ICL without kNN search can achieve a comparable performance with strong supervised models on the TOP tasks, and 2) kNN-ICL significantly improves the comprehension of complex requests by seamlessly integrating ICL with a nearest-neighbor approach. Notably, this enhancement is achieved without the need for additional data or specialized prompts.