@inproceedings{he-etal-2024-never,
title = "Never Lost in the Middle: Mastering Long-Context Question Answering with Position-Agnostic Decompositional Training",
author = "He, Junqing and
Pan, Kunhao and
Dong, Xiaoqun and
Song, Zhuoyang and
LiuYiBo, LiuYiBo and
Qianguosun, Qianguosun and
Liang, Yuxin and
Wang, Hao and
Zhang, Enming and
Zhang, Jiaxing",
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.736/",
doi = "10.18653/v1/2024.acl-long.736",
pages = "13628--13642",
abstract = "While large language models (LLMs) are equipped with longer text input capabilities than before, they are struggling to seek correct information in long contexts. The {\textquotedblleft}lost in the middle{\textquotedblright} problem challenges most LLMs, referring to the dramatic decline in accuracy when correct information is located in the middle. To overcome this crucial issue, this paper proposes to enhance the information searching and reflection ability of LLMs in long contexts via specially designed tasks called Position-Agnostic Multi-step QA (PAM QA). Trained in this task, our model excels in focusing more precisely on the desired information. Experimental results show substantial improvement in Multi-doc QA and other benchmarks, superior to state-of-the-art models by 13.7{\%} absolute gain in shuffled settings, by 21.5{\%} in passage retrieval task. We release our model and code to promote related research in the community."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="he-etal-2024-never">
<titleInfo>
<title>Never Lost in the Middle: Mastering Long-Context Question Answering with Position-Agnostic Decompositional Training</title>
</titleInfo>
<name type="personal">
<namePart type="given">Junqing</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kunhao</namePart>
<namePart type="family">Pan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaoqun</namePart>
<namePart type="family">Dong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhuoyang</namePart>
<namePart type="family">Song</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">LiuYiBo</namePart>
<namePart type="family">LiuYiBo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qianguosun</namePart>
<namePart type="family">Qianguosun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuxin</namePart>
<namePart type="family">Liang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hao</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Enming</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiaxing</namePart>
<namePart type="family">Zhang</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>While large language models (LLMs) are equipped with longer text input capabilities than before, they are struggling to seek correct information in long contexts. The “lost in the middle” problem challenges most LLMs, referring to the dramatic decline in accuracy when correct information is located in the middle. To overcome this crucial issue, this paper proposes to enhance the information searching and reflection ability of LLMs in long contexts via specially designed tasks called Position-Agnostic Multi-step QA (PAM QA). Trained in this task, our model excels in focusing more precisely on the desired information. Experimental results show substantial improvement in Multi-doc QA and other benchmarks, superior to state-of-the-art models by 13.7% absolute gain in shuffled settings, by 21.5% in passage retrieval task. We release our model and code to promote related research in the community.</abstract>
<identifier type="citekey">he-etal-2024-never</identifier>
<identifier type="doi">10.18653/v1/2024.acl-long.736</identifier>
<location>
<url>https://aclanthology.org/2024.luhme-long.736/</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>13628</start>
<end>13642</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Never Lost in the Middle: Mastering Long-Context Question Answering with Position-Agnostic Decompositional Training
%A He, Junqing
%A Pan, Kunhao
%A Dong, Xiaoqun
%A Song, Zhuoyang
%A LiuYiBo, LiuYiBo
%A Qianguosun, Qianguosun
%A Liang, Yuxin
%A Wang, Hao
%A Zhang, Enming
%A Zhang, Jiaxing
%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 he-etal-2024-never
%X While large language models (LLMs) are equipped with longer text input capabilities than before, they are struggling to seek correct information in long contexts. The “lost in the middle” problem challenges most LLMs, referring to the dramatic decline in accuracy when correct information is located in the middle. To overcome this crucial issue, this paper proposes to enhance the information searching and reflection ability of LLMs in long contexts via specially designed tasks called Position-Agnostic Multi-step QA (PAM QA). Trained in this task, our model excels in focusing more precisely on the desired information. Experimental results show substantial improvement in Multi-doc QA and other benchmarks, superior to state-of-the-art models by 13.7% absolute gain in shuffled settings, by 21.5% in passage retrieval task. We release our model and code to promote related research in the community.
%R 10.18653/v1/2024.acl-long.736
%U https://aclanthology.org/2024.luhme-long.736/
%U https://doi.org/10.18653/v1/2024.acl-long.736
%P 13628-13642
Markdown (Informal)
[Never Lost in the Middle: Mastering Long-Context Question Answering with Position-Agnostic Decompositional Training](https://aclanthology.org/2024.luhme-long.736/) (He et al., ACL 2024)
ACL
- Junqing He, Kunhao Pan, Xiaoqun Dong, Zhuoyang Song, LiuYiBo LiuYiBo, Qianguosun Qianguosun, Yuxin Liang, Hao Wang, Enming Zhang, and Jiaxing Zhang. 2024. Never Lost in the Middle: Mastering Long-Context Question Answering with Position-Agnostic Decompositional Training. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 13628–13642, Bangkok, Thailand. Association for Computational Linguistics.