@inproceedings{thirukovalluru-etal-2024-atomic,
title = "Atomic Self-Consistency for Better Long Form Generations",
author = "Thirukovalluru, Raghuveer and
Huang, Yukun and
Dhingra, Bhuwan",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.706/",
doi = "10.18653/v1/2024.emnlp-main.706",
pages = "12681--12694",
abstract = "Recent work has aimed to improve LLM generations by filtering out hallucinations, thereby improving the precision of the information in responses. Correctness of a long-form response, however, also depends on the recall of multiple pieces of information relevant to the question. In this paper, we introduce Atomic Self-Consistency (ASC), a technique for improving the recall of relevant information in an LLM response. ASC follows recent work, Universal Self-Consistency (USC) in using multiple stochastic samples from an LLM to improve the long-form response. Unlike USC which only focuses on selecting the best single generation, ASC picks authentic subparts from the samples and merges them into a superior composite answer. Through extensive experiments and ablations, we show that merging relevant subparts of multiple samples performs significantly better than picking a single sample. ASC demonstrates significant gains over USC on multiple factoids and open-ended QA datasets - ASQA, QAMPARI, QUEST, ELI5 with ChatGPT and Llama3. Our analysis also reveals untapped potential for enhancing long-form generations using the approach of merging multiple samples."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="thirukovalluru-etal-2024-atomic">
<titleInfo>
<title>Atomic Self-Consistency for Better Long Form Generations</title>
</titleInfo>
<name type="personal">
<namePart type="given">Raghuveer</namePart>
<namePart type="family">Thirukovalluru</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yukun</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bhuwan</namePart>
<namePart type="family">Dhingra</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 2024 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yaser</namePart>
<namePart type="family">Al-Onaizan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohit</namePart>
<namePart type="family">Bansal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yun-Nung</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Miami, Florida, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Recent work has aimed to improve LLM generations by filtering out hallucinations, thereby improving the precision of the information in responses. Correctness of a long-form response, however, also depends on the recall of multiple pieces of information relevant to the question. In this paper, we introduce Atomic Self-Consistency (ASC), a technique for improving the recall of relevant information in an LLM response. ASC follows recent work, Universal Self-Consistency (USC) in using multiple stochastic samples from an LLM to improve the long-form response. Unlike USC which only focuses on selecting the best single generation, ASC picks authentic subparts from the samples and merges them into a superior composite answer. Through extensive experiments and ablations, we show that merging relevant subparts of multiple samples performs significantly better than picking a single sample. ASC demonstrates significant gains over USC on multiple factoids and open-ended QA datasets - ASQA, QAMPARI, QUEST, ELI5 with ChatGPT and Llama3. Our analysis also reveals untapped potential for enhancing long-form generations using the approach of merging multiple samples.</abstract>
<identifier type="citekey">thirukovalluru-etal-2024-atomic</identifier>
<identifier type="doi">10.18653/v1/2024.emnlp-main.706</identifier>
<location>
<url>https://aclanthology.org/2024.emnlp-main.706/</url>
</location>
<part>
<date>2024-11</date>
<extent unit="page">
<start>12681</start>
<end>12694</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Atomic Self-Consistency for Better Long Form Generations
%A Thirukovalluru, Raghuveer
%A Huang, Yukun
%A Dhingra, Bhuwan
%Y Al-Onaizan, Yaser
%Y Bansal, Mohit
%Y Chen, Yun-Nung
%S Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
%D 2024
%8 November
%I Association for Computational Linguistics
%C Miami, Florida, USA
%F thirukovalluru-etal-2024-atomic
%X Recent work has aimed to improve LLM generations by filtering out hallucinations, thereby improving the precision of the information in responses. Correctness of a long-form response, however, also depends on the recall of multiple pieces of information relevant to the question. In this paper, we introduce Atomic Self-Consistency (ASC), a technique for improving the recall of relevant information in an LLM response. ASC follows recent work, Universal Self-Consistency (USC) in using multiple stochastic samples from an LLM to improve the long-form response. Unlike USC which only focuses on selecting the best single generation, ASC picks authentic subparts from the samples and merges them into a superior composite answer. Through extensive experiments and ablations, we show that merging relevant subparts of multiple samples performs significantly better than picking a single sample. ASC demonstrates significant gains over USC on multiple factoids and open-ended QA datasets - ASQA, QAMPARI, QUEST, ELI5 with ChatGPT and Llama3. Our analysis also reveals untapped potential for enhancing long-form generations using the approach of merging multiple samples.
%R 10.18653/v1/2024.emnlp-main.706
%U https://aclanthology.org/2024.emnlp-main.706/
%U https://doi.org/10.18653/v1/2024.emnlp-main.706
%P 12681-12694
Markdown (Informal)
[Atomic Self-Consistency for Better Long Form Generations](https://aclanthology.org/2024.emnlp-main.706/) (Thirukovalluru et al., EMNLP 2024)
ACL
- Raghuveer Thirukovalluru, Yukun Huang, and Bhuwan Dhingra. 2024. Atomic Self-Consistency for Better Long Form Generations. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 12681–12694, Miami, Florida, USA. Association for Computational Linguistics.