Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Zijie, Rashid, Farzana, Blanco, Eduardo
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2404.16262
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917649929207808
author Wang, Zijie
Rashid, Farzana
Blanco, Eduardo
author_facet Wang, Zijie
Rashid, Farzana
Blanco, Eduardo
contents People often answer yes-no questions without explicitly saying yes, no, or similar polar keywords. Figuring out the meaning of indirect answers is challenging, even for large language models. In this paper, we investigate this problem working with dialogues from multiple domains. We present new benchmarks in three diverse domains: movie scripts, tennis interviews, and airline customer service. We present an approach grounded on distant supervision and blended training to quickly adapt to a new dialogue domain. Experimental results show that our approach is never detrimental and yields F1 improvements as high as 11-34%.
format Preprint
id arxiv_https___arxiv_org_abs_2404_16262
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Interpreting Answers to Yes-No Questions in Dialogues from Multiple Domains
Wang, Zijie
Rashid, Farzana
Blanco, Eduardo
Computation and Language
People often answer yes-no questions without explicitly saying yes, no, or similar polar keywords. Figuring out the meaning of indirect answers is challenging, even for large language models. In this paper, we investigate this problem working with dialogues from multiple domains. We present new benchmarks in three diverse domains: movie scripts, tennis interviews, and airline customer service. We present an approach grounded on distant supervision and blended training to quickly adapt to a new dialogue domain. Experimental results show that our approach is never detrimental and yields F1 improvements as high as 11-34%.
title Interpreting Answers to Yes-No Questions in Dialogues from Multiple Domains
topic Computation and Language
url https://arxiv.org/abs/2404.16262