Leveraging Code to Improve In-context Learning for Semantic Parsing

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Bogin, Ben, Gupta, Shivanshu, Clark, Peter, Sabharwal, Ashish
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911817502031872
author Bogin, Ben
Gupta, Shivanshu
Clark, Peter
Sabharwal, Ashish
author_facet Bogin, Ben
Gupta, Shivanshu
Clark, Peter
Sabharwal, Ashish
contents In-context learning (ICL) is an appealing approach for semantic parsing due to its few-shot nature and improved generalization. However, learning to parse to rare domain-specific languages (DSLs) from just a few demonstrations is challenging, limiting the performance of even the most capable LLMs. In this work, we improve the effectiveness of ICL for semantic parsing by (1) using general-purpose programming languages such as Python instead of DSLs, and (2) augmenting prompts with a structured domain description that includes, e.g., the available classes and functions. We show that both these changes significantly improve accuracy across three popular datasets. Combined, they lead to dramatic improvements (e.g. 7.9% to 66.5% on SMCalFlow compositional split), nearly closing the performance gap between easier i.i.d.\ and harder compositional splits when used with a strong model, and reducing the need for a large number of demonstrations. We find that the resemblance of the target parse language to general-purpose code is a more important factor than the language's popularity in pre-training corpora. Our findings provide an improved methodology for building semantic parsers in the modern context of ICL with LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2311_09519
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Leveraging Code to Improve In-context Learning for Semantic Parsing
Bogin, Ben
Gupta, Shivanshu
Clark, Peter
Sabharwal, Ashish
Computation and Language
In-context learning (ICL) is an appealing approach for semantic parsing due to its few-shot nature and improved generalization. However, learning to parse to rare domain-specific languages (DSLs) from just a few demonstrations is challenging, limiting the performance of even the most capable LLMs. In this work, we improve the effectiveness of ICL for semantic parsing by (1) using general-purpose programming languages such as Python instead of DSLs, and (2) augmenting prompts with a structured domain description that includes, e.g., the available classes and functions. We show that both these changes significantly improve accuracy across three popular datasets. Combined, they lead to dramatic improvements (e.g. 7.9% to 66.5% on SMCalFlow compositional split), nearly closing the performance gap between easier i.i.d.\ and harder compositional splits when used with a strong model, and reducing the need for a large number of demonstrations. We find that the resemblance of the target parse language to general-purpose code is a more important factor than the language's popularity in pre-training corpora. Our findings provide an improved methodology for building semantic parsers in the modern context of ICL with LLMs.
title Leveraging Code to Improve In-context Learning for Semantic Parsing
topic Computation and Language
url https://arxiv.org/abs/2311.09519