@inproceedings{huang-etal-2024-characterizing,
title = "Characterizing Similarities and Divergences in Conversational Tones in Humans and {LLM}s by Sampling with People",
author = "Huang, Dun-Ming and
Van Rijn, Pol and
Sucholutsky, Ilia and
Marjieh, Raja and
Jacoby, Nori",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.luhme-long.565/",
doi = "10.18653/v1/2024.acl-long.565",
pages = "10486--10512",
abstract = "Conversational tones {---} the manners and attitudes in which speakers communicate {---} are essential to effective communication. As Large Language Models (LLMs) become increasingly popular, it is necessary to characterize the divergences in their conversational tones relative to humans. Prior research relied on pre-existing taxonomies or text corpora, which suffer from experimenter bias and may not be representative of real-world distributions. Inspired by methods from cognitive science, we propose an iterative method for simultaneously eliciting conversational tones and sentences, where participants alternate between two tasks: (1) one participant identifies the tone of a given sentence and (2) a different participant generates a sentence based on that tone. We run 50 iterations of this process with both human participants and GPT-4 and obtain a dataset of sentences and frequent conversational tones. In an additional experiment, humans and GPT-4 annotated all sentences with all tones. With data from 1,339 participants, 33,370 human judgments, and 29,900 GPT-4 queries, we show how our approach can be used to create an interpretable geometric representation of relations between tones in humans and GPT-4. This work showcases how combining ideas from machine learning and cognitive science can address challenges in human-computer interactions."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="huang-etal-2024-characterizing">
<titleInfo>
<title>Characterizing Similarities and Divergences in Conversational Tones in Humans and LLMs by Sampling with People</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dun-Ming</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pol</namePart>
<namePart type="family">Van Rijn</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ilia</namePart>
<namePart type="family">Sucholutsky</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Raja</namePart>
<namePart type="family">Marjieh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nori</namePart>
<namePart type="family">Jacoby</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lun-Wei</namePart>
<namePart type="family">Ku</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andre</namePart>
<namePart type="family">Martins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivek</namePart>
<namePart type="family">Srikumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Conversational tones — the manners and attitudes in which speakers communicate — are essential to effective communication. As Large Language Models (LLMs) become increasingly popular, it is necessary to characterize the divergences in their conversational tones relative to humans. Prior research relied on pre-existing taxonomies or text corpora, which suffer from experimenter bias and may not be representative of real-world distributions. Inspired by methods from cognitive science, we propose an iterative method for simultaneously eliciting conversational tones and sentences, where participants alternate between two tasks: (1) one participant identifies the tone of a given sentence and (2) a different participant generates a sentence based on that tone. We run 50 iterations of this process with both human participants and GPT-4 and obtain a dataset of sentences and frequent conversational tones. In an additional experiment, humans and GPT-4 annotated all sentences with all tones. With data from 1,339 participants, 33,370 human judgments, and 29,900 GPT-4 queries, we show how our approach can be used to create an interpretable geometric representation of relations between tones in humans and GPT-4. This work showcases how combining ideas from machine learning and cognitive science can address challenges in human-computer interactions.</abstract>
<identifier type="citekey">huang-etal-2024-characterizing</identifier>
<identifier type="doi">10.18653/v1/2024.acl-long.565</identifier>
<location>
<url>https://aclanthology.org/2024.luhme-long.565/</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>10486</start>
<end>10512</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Characterizing Similarities and Divergences in Conversational Tones in Humans and LLMs by Sampling with People
%A Huang, Dun-Ming
%A Van Rijn, Pol
%A Sucholutsky, Ilia
%A Marjieh, Raja
%A Jacoby, Nori
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F huang-etal-2024-characterizing
%X Conversational tones — the manners and attitudes in which speakers communicate — are essential to effective communication. As Large Language Models (LLMs) become increasingly popular, it is necessary to characterize the divergences in their conversational tones relative to humans. Prior research relied on pre-existing taxonomies or text corpora, which suffer from experimenter bias and may not be representative of real-world distributions. Inspired by methods from cognitive science, we propose an iterative method for simultaneously eliciting conversational tones and sentences, where participants alternate between two tasks: (1) one participant identifies the tone of a given sentence and (2) a different participant generates a sentence based on that tone. We run 50 iterations of this process with both human participants and GPT-4 and obtain a dataset of sentences and frequent conversational tones. In an additional experiment, humans and GPT-4 annotated all sentences with all tones. With data from 1,339 participants, 33,370 human judgments, and 29,900 GPT-4 queries, we show how our approach can be used to create an interpretable geometric representation of relations between tones in humans and GPT-4. This work showcases how combining ideas from machine learning and cognitive science can address challenges in human-computer interactions.
%R 10.18653/v1/2024.acl-long.565
%U https://aclanthology.org/2024.luhme-long.565/
%U https://doi.org/10.18653/v1/2024.acl-long.565
%P 10486-10512
Markdown (Informal)
[Characterizing Similarities and Divergences in Conversational Tones in Humans and LLMs by Sampling with People](https://aclanthology.org/2024.luhme-long.565/) (Huang et al., ACL 2024)
ACL