@inproceedings{zhou-etal-2024-hypothesis,
title = "Hypothesis Generation with Large Language Models",
author = "Zhou, Yangqiaoyu and
Liu, Haokun and
Srivastava, Tejes and
Mei, Hongyuan and
Tan, Chenhao",
editor = "Peled-Cohen, Lotem and
Calderon, Nitay and
Lissak, Shir and
Reichart, Roi",
booktitle = "Proceedings of the 1st Workshop on NLP for Science (NLP4Science)",
month = nov,
year = "2024",
address = "Miami, FL, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.nlp4science-1.10",
pages = "117--139",
abstract = "Effective generation of novel hypotheses is instrumental to scientific progress. So far, researchers have been the main powerhouse behind hypothesis generation by painstaking data analysis and thinking (also known as the Eureka moment). In this paper, we examine the potential of large language models (LLMs) to generate hypotheses. We focus on hypothesis generation based on data (i.e., labeled examples). To enable LLMs to handle Long contexts, we generate initial hypotheses from a small number of examples and then update them iteratively to improve the quality of hypotheses. Inspired by multi-armed bandits, we design a reward function to inform the exploitation-exploration tradeoff in the update process. Our algorithm is able to generate hypotheses that enable much better predictive performance than few-shot prompting in classification tasks, improving accuracy by 31.7{\%} on a synthetic dataset and by 13.9{\%}, 3.3{\%} and, 24.9{\%} on three real-world datasets. We also outperform supervised learning by 12.1{\%} and 11.6{\%} on two challenging real-world datasets. Furthermore, we find that the generated hypotheses not only corroborate human-verified theories but also uncover new insights for the tasks.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhou-etal-2024-hypothesis">
<titleInfo>
<title>Hypothesis Generation with Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yangqiaoyu</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haokun</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tejes</namePart>
<namePart type="family">Srivastava</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hongyuan</namePart>
<namePart type="family">Mei</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chenhao</namePart>
<namePart type="family">Tan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 1st Workshop on NLP for Science (NLP4Science)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lotem</namePart>
<namePart type="family">Peled-Cohen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nitay</namePart>
<namePart type="family">Calderon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shir</namePart>
<namePart type="family">Lissak</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Roi</namePart>
<namePart type="family">Reichart</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Miami, FL, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Effective generation of novel hypotheses is instrumental to scientific progress. So far, researchers have been the main powerhouse behind hypothesis generation by painstaking data analysis and thinking (also known as the Eureka moment). In this paper, we examine the potential of large language models (LLMs) to generate hypotheses. We focus on hypothesis generation based on data (i.e., labeled examples). To enable LLMs to handle Long contexts, we generate initial hypotheses from a small number of examples and then update them iteratively to improve the quality of hypotheses. Inspired by multi-armed bandits, we design a reward function to inform the exploitation-exploration tradeoff in the update process. Our algorithm is able to generate hypotheses that enable much better predictive performance than few-shot prompting in classification tasks, improving accuracy by 31.7% on a synthetic dataset and by 13.9%, 3.3% and, 24.9% on three real-world datasets. We also outperform supervised learning by 12.1% and 11.6% on two challenging real-world datasets. Furthermore, we find that the generated hypotheses not only corroborate human-verified theories but also uncover new insights for the tasks.</abstract>
<identifier type="citekey">zhou-etal-2024-hypothesis</identifier>
<location>
<url>https://aclanthology.org/2024.nlp4science-1.10</url>
</location>
<part>
<date>2024-11</date>
<extent unit="page">
<start>117</start>
<end>139</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Hypothesis Generation with Large Language Models
%A Zhou, Yangqiaoyu
%A Liu, Haokun
%A Srivastava, Tejes
%A Mei, Hongyuan
%A Tan, Chenhao
%Y Peled-Cohen, Lotem
%Y Calderon, Nitay
%Y Lissak, Shir
%Y Reichart, Roi
%S Proceedings of the 1st Workshop on NLP for Science (NLP4Science)
%D 2024
%8 November
%I Association for Computational Linguistics
%C Miami, FL, USA
%F zhou-etal-2024-hypothesis
%X Effective generation of novel hypotheses is instrumental to scientific progress. So far, researchers have been the main powerhouse behind hypothesis generation by painstaking data analysis and thinking (also known as the Eureka moment). In this paper, we examine the potential of large language models (LLMs) to generate hypotheses. We focus on hypothesis generation based on data (i.e., labeled examples). To enable LLMs to handle Long contexts, we generate initial hypotheses from a small number of examples and then update them iteratively to improve the quality of hypotheses. Inspired by multi-armed bandits, we design a reward function to inform the exploitation-exploration tradeoff in the update process. Our algorithm is able to generate hypotheses that enable much better predictive performance than few-shot prompting in classification tasks, improving accuracy by 31.7% on a synthetic dataset and by 13.9%, 3.3% and, 24.9% on three real-world datasets. We also outperform supervised learning by 12.1% and 11.6% on two challenging real-world datasets. Furthermore, we find that the generated hypotheses not only corroborate human-verified theories but also uncover new insights for the tasks.
%U https://aclanthology.org/2024.nlp4science-1.10
%P 117-139
Markdown (Informal)
[Hypothesis Generation with Large Language Models](https://aclanthology.org/2024.nlp4science-1.10) (Zhou et al., NLP4Science 2024)
ACL
- Yangqiaoyu Zhou, Haokun Liu, Tejes Srivastava, Hongyuan Mei, and Chenhao Tan. 2024. Hypothesis Generation with Large Language Models. In Proceedings of the 1st Workshop on NLP for Science (NLP4Science), pages 117–139, Miami, FL, USA. Association for Computational Linguistics.