@inproceedings{yuan-etal-2023-selecting,
title = "Selecting Better Samples from Pre-trained {LLM}s: A Case Study on Question Generation",
author = "Yuan, Xingdi and
Wang, Tong and
Wang, Yen-Hsiang and
Fine, Emery and
Abdelghani, Rania and
Sauz{\'e}on, H{\'e}l{\`e}ne and
Oudeyer, Pierre-Yves",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2023",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.findings-acl.820",
doi = "10.18653/v1/2023.findings-acl.820",
pages = "12952--12965",
abstract = "Large Language Models (LLMs) have in recent years demonstrated impressive prowess in natural language generation. A common practice to improve generation diversity is to sample multiple outputs from the model. However, partly due to the inaccessibility of LLMs, there lacks a simple and robust way of selecting the best output from these stochastic samples. As a case study framed in the context of question generation, we propose two prompt-based approaches, namely round-trip and prompt-based score, to selecting high-quality questions from a set of LLM-generated candidates. Our method works without the need to modify the underlying model, nor does it rely on human-annotated references {---} both of which are realistic constraints for real-world deployment of LLMs. With automatic as well as human evaluations, we empirically demonstrate that our approach can effectively select questions of higher qualities than greedy generation.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yuan-etal-2023-selecting">
<titleInfo>
<title>Selecting Better Samples from Pre-trained LLMs: A Case Study on Question Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xingdi</namePart>
<namePart type="family">Yuan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yen-Hsiang</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Emery</namePart>
<namePart type="family">Fine</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rania</namePart>
<namePart type="family">Abdelghani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hélène</namePart>
<namePart type="family">Sauzéon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pierre-Yves</namePart>
<namePart type="family">Oudeyer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2023</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rogers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Boyd-Graber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naoaki</namePart>
<namePart type="family">Okazaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Large Language Models (LLMs) have in recent years demonstrated impressive prowess in natural language generation. A common practice to improve generation diversity is to sample multiple outputs from the model. However, partly due to the inaccessibility of LLMs, there lacks a simple and robust way of selecting the best output from these stochastic samples. As a case study framed in the context of question generation, we propose two prompt-based approaches, namely round-trip and prompt-based score, to selecting high-quality questions from a set of LLM-generated candidates. Our method works without the need to modify the underlying model, nor does it rely on human-annotated references — both of which are realistic constraints for real-world deployment of LLMs. With automatic as well as human evaluations, we empirically demonstrate that our approach can effectively select questions of higher qualities than greedy generation.</abstract>
<identifier type="citekey">yuan-etal-2023-selecting</identifier>
<identifier type="doi">10.18653/v1/2023.findings-acl.820</identifier>
<location>
<url>https://aclanthology.org/2023.findings-acl.820</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>12952</start>
<end>12965</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Selecting Better Samples from Pre-trained LLMs: A Case Study on Question Generation
%A Yuan, Xingdi
%A Wang, Tong
%A Wang, Yen-Hsiang
%A Fine, Emery
%A Abdelghani, Rania
%A Sauzéon, Hélène
%A Oudeyer, Pierre-Yves
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Findings of the Association for Computational Linguistics: ACL 2023
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F yuan-etal-2023-selecting
%X Large Language Models (LLMs) have in recent years demonstrated impressive prowess in natural language generation. A common practice to improve generation diversity is to sample multiple outputs from the model. However, partly due to the inaccessibility of LLMs, there lacks a simple and robust way of selecting the best output from these stochastic samples. As a case study framed in the context of question generation, we propose two prompt-based approaches, namely round-trip and prompt-based score, to selecting high-quality questions from a set of LLM-generated candidates. Our method works without the need to modify the underlying model, nor does it rely on human-annotated references — both of which are realistic constraints for real-world deployment of LLMs. With automatic as well as human evaluations, we empirically demonstrate that our approach can effectively select questions of higher qualities than greedy generation.
%R 10.18653/v1/2023.findings-acl.820
%U https://aclanthology.org/2023.findings-acl.820
%U https://doi.org/10.18653/v1/2023.findings-acl.820
%P 12952-12965
Markdown (Informal)
[Selecting Better Samples from Pre-trained LLMs: A Case Study on Question Generation](https://aclanthology.org/2023.findings-acl.820) (Yuan et al., Findings 2023)
ACL