@inproceedings{huang-etal-2024-learning,
title = "Learning Fine-Grained Grounded Citations for Attributed Large Language Models",
author = "Huang, Lei and
Feng, Xiaocheng and
Ma, Weitao and
Gu, Yuxuan and
Zhong, Weihong and
Feng, Xiachong and
Yu, Weijiang and
Peng, Weihua and
Tang, Duyu and
Tu, Dandan and
Qin, Bing",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2024",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-acl.838",
doi = "10.18653/v1/2024.findings-acl.838",
pages = "14095--14113",
abstract = "Despite the impressive performance on information-seeking tasks, large language models (LLMs) still struggle with hallucinations. Attributed LLMs, which augment generated text with in-line citations, demonstrate potential in mitigating hallucinations and improving verifiability. However, current approaches suffer from suboptimal citation quality due to their reliance on in-context learning. Furthermore, the practice of merely citing document identifiers complicates the process for users to pinpoint specific supporting evidence. In this work, we introduce FRONT, a training framework that teaches LLMs to generate Fine-grained grounded citations. By initially grounding fine-grained supporting quotes, which then guide the generation process, these quotes not only provide supervision signals to improve citation quality but also serve as fine-grained attributions. Experiments on the ALCE benchmark demonstrate the efficacy of FRONT in generating superior grounded responses and highly supportive citations. With LLaMA-2-7B, the framework significantly outperforms all the baselines, achieving an average of 14.21{\%} improvement in citation quality across all datasets, even surpassing ChatGPT.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="huang-etal-2024-learning">
<titleInfo>
<title>Learning Fine-Grained Grounded Citations for Attributed Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lei</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaocheng</namePart>
<namePart type="family">Feng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weitao</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuxuan</namePart>
<namePart type="family">Gu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weihong</namePart>
<namePart type="family">Zhong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiachong</namePart>
<namePart type="family">Feng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weijiang</namePart>
<namePart type="family">Yu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weihua</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Duyu</namePart>
<namePart type="family">Tang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dandan</namePart>
<namePart type="family">Tu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bing</namePart>
<namePart type="family">Qin</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>Findings of the Association for Computational Linguistics: ACL 2024</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>Despite the impressive performance on information-seeking tasks, large language models (LLMs) still struggle with hallucinations. Attributed LLMs, which augment generated text with in-line citations, demonstrate potential in mitigating hallucinations and improving verifiability. However, current approaches suffer from suboptimal citation quality due to their reliance on in-context learning. Furthermore, the practice of merely citing document identifiers complicates the process for users to pinpoint specific supporting evidence. In this work, we introduce FRONT, a training framework that teaches LLMs to generate Fine-grained grounded citations. By initially grounding fine-grained supporting quotes, which then guide the generation process, these quotes not only provide supervision signals to improve citation quality but also serve as fine-grained attributions. Experiments on the ALCE benchmark demonstrate the efficacy of FRONT in generating superior grounded responses and highly supportive citations. With LLaMA-2-7B, the framework significantly outperforms all the baselines, achieving an average of 14.21% improvement in citation quality across all datasets, even surpassing ChatGPT.</abstract>
<identifier type="citekey">huang-etal-2024-learning</identifier>
<identifier type="doi">10.18653/v1/2024.findings-acl.838</identifier>
<location>
<url>https://aclanthology.org/2024.findings-acl.838</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>14095</start>
<end>14113</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Learning Fine-Grained Grounded Citations for Attributed Large Language Models
%A Huang, Lei
%A Feng, Xiaocheng
%A Ma, Weitao
%A Gu, Yuxuan
%A Zhong, Weihong
%A Feng, Xiachong
%A Yu, Weijiang
%A Peng, Weihua
%A Tang, Duyu
%A Tu, Dandan
%A Qin, Bing
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Findings of the Association for Computational Linguistics: ACL 2024
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F huang-etal-2024-learning
%X Despite the impressive performance on information-seeking tasks, large language models (LLMs) still struggle with hallucinations. Attributed LLMs, which augment generated text with in-line citations, demonstrate potential in mitigating hallucinations and improving verifiability. However, current approaches suffer from suboptimal citation quality due to their reliance on in-context learning. Furthermore, the practice of merely citing document identifiers complicates the process for users to pinpoint specific supporting evidence. In this work, we introduce FRONT, a training framework that teaches LLMs to generate Fine-grained grounded citations. By initially grounding fine-grained supporting quotes, which then guide the generation process, these quotes not only provide supervision signals to improve citation quality but also serve as fine-grained attributions. Experiments on the ALCE benchmark demonstrate the efficacy of FRONT in generating superior grounded responses and highly supportive citations. With LLaMA-2-7B, the framework significantly outperforms all the baselines, achieving an average of 14.21% improvement in citation quality across all datasets, even surpassing ChatGPT.
%R 10.18653/v1/2024.findings-acl.838
%U https://aclanthology.org/2024.findings-acl.838
%U https://doi.org/10.18653/v1/2024.findings-acl.838
%P 14095-14113
Markdown (Informal)
[Learning Fine-Grained Grounded Citations for Attributed Large Language Models](https://aclanthology.org/2024.findings-acl.838) (Huang et al., Findings 2024)
ACL
- Lei Huang, Xiaocheng Feng, Weitao Ma, Yuxuan Gu, Weihong Zhong, Xiachong Feng, Weijiang Yu, Weihua Peng, Duyu Tang, Dandan Tu, and Bing Qin. 2024. Learning Fine-Grained Grounded Citations for Attributed Large Language Models. In Findings of the Association for Computational Linguistics: ACL 2024, pages 14095–14113, Bangkok, Thailand. Association for Computational Linguistics.