Generating Demonstrations for In-Context Compositional Generalization in Grounded Language Learning

Loading...
Thumbnail Image

Access rights

openAccess
publishedVersion

URL

Journal Title

Journal ISSN

Volume Title

A4 Artikkeli konferenssijulkaisussa

Major/Subject

Mcode

Degree programme

Language

en

Pages

Series

Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pp. 15960–15991

Abstract

In-Context-learning and few-shot prompting are viable methods compositional output generation. However, these methods can be very sensitive to the choice of support examples used. Retrieving good supports from the training data for a given test query is already a difficult problem, but in some cases solving this may not even be enough. We consider the setting of grounded language learning problems where finding relevant supports in the same or similar states as the query may be difficult. We design an agent which instead generates possible supports inputs and targets current state of the world, then uses them in-context-learning to solve the test query. We show substantially improved performance on a previously unsolved compositional generalization test without a loss of performance in other areas. The approach is general and can even scale to instructions expressed in natural language.

Description

Keywords

Other note

Citation

Spilsbury, S, Marttinen, P & Ilin, A 2024, Generating Demonstrations for In-Context Compositional Generalization in Grounded Language Learning. in Y Al-Onaizan, M Bansal & Y-N Chen (eds), Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics, pp. 15960–15991, Conference on Empirical Methods in Natural Language Processing, Miami, Florida, United States, 12/11/2024. https://doi.org/10.18653/v1/2024.emnlp-main.893