@inproceedings{chen-etal-2025-augment,
title = "When and How to Augment Your Input: Question Routing Helps Balance the Accuracy and Efficiency of Large Language Models",
author = "Chen, Shufan and
Zheng, He and
Cui, Lei",
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Findings of the Association for Computational Linguistics: NAACL 2025",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-naacl.200/",
doi = "10.18653/v1/2025.findings-naacl.200",
pages = "3621--3634",
ISBN = "979-8-89176-195-7",
abstract = "Although large language models rely on parametric knowledge to achieve exceptional performance across various question-answering tasks, they still face challenges when addressing knowledge-based long-tail questions. Augmented generation techniques, such as chain-of-thought prompting and retrieval augmentation, can effectively enhance the ability of these models to answer long-tail questions. However, improving accuracy through augmented generation often results in significant latency within question-answering systems. This paper addresses the issue of ``when and how to augment the input'' by proposing an adaptive question routing framework. This framework employs a query router to select the most appropriate augmentation path at the right time, thereby enhancing both the accuracy and efficiency of question-answering systems. Extensive comparative experiments on benchmarks such as AmbigNQ, HotpotQA, MMLU-STEM, and PopQA demonstrate that our method surpasses existing approaches in both accuracy and efficiency. Furthermore, this paper introduces two metrics for evaluating adaptive question augmentation methods and presents a new benchmark for adaptive question augmentation, aiming to advance the field."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chen-etal-2025-augment">
<titleInfo>
<title>When and How to Augment Your Input: Question Routing Helps Balance the Accuracy and Efficiency of Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Shufan</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">He</namePart>
<namePart type="family">Zheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lei</namePart>
<namePart type="family">Cui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: NAACL 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luis</namePart>
<namePart type="family">Chiruzzo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ritter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-195-7</identifier>
</relatedItem>
<abstract>Although large language models rely on parametric knowledge to achieve exceptional performance across various question-answering tasks, they still face challenges when addressing knowledge-based long-tail questions. Augmented generation techniques, such as chain-of-thought prompting and retrieval augmentation, can effectively enhance the ability of these models to answer long-tail questions. However, improving accuracy through augmented generation often results in significant latency within question-answering systems. This paper addresses the issue of “when and how to augment the input” by proposing an adaptive question routing framework. This framework employs a query router to select the most appropriate augmentation path at the right time, thereby enhancing both the accuracy and efficiency of question-answering systems. Extensive comparative experiments on benchmarks such as AmbigNQ, HotpotQA, MMLU-STEM, and PopQA demonstrate that our method surpasses existing approaches in both accuracy and efficiency. Furthermore, this paper introduces two metrics for evaluating adaptive question augmentation methods and presents a new benchmark for adaptive question augmentation, aiming to advance the field.</abstract>
<identifier type="citekey">chen-etal-2025-augment</identifier>
<identifier type="doi">10.18653/v1/2025.findings-naacl.200</identifier>
<location>
<url>https://aclanthology.org/2025.findings-naacl.200/</url>
</location>
<part>
<date>2025-04</date>
<extent unit="page">
<start>3621</start>
<end>3634</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T When and How to Augment Your Input: Question Routing Helps Balance the Accuracy and Efficiency of Large Language Models
%A Chen, Shufan
%A Zheng, He
%A Cui, Lei
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Findings of the Association for Computational Linguistics: NAACL 2025
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-195-7
%F chen-etal-2025-augment
%X Although large language models rely on parametric knowledge to achieve exceptional performance across various question-answering tasks, they still face challenges when addressing knowledge-based long-tail questions. Augmented generation techniques, such as chain-of-thought prompting and retrieval augmentation, can effectively enhance the ability of these models to answer long-tail questions. However, improving accuracy through augmented generation often results in significant latency within question-answering systems. This paper addresses the issue of “when and how to augment the input” by proposing an adaptive question routing framework. This framework employs a query router to select the most appropriate augmentation path at the right time, thereby enhancing both the accuracy and efficiency of question-answering systems. Extensive comparative experiments on benchmarks such as AmbigNQ, HotpotQA, MMLU-STEM, and PopQA demonstrate that our method surpasses existing approaches in both accuracy and efficiency. Furthermore, this paper introduces two metrics for evaluating adaptive question augmentation methods and presents a new benchmark for adaptive question augmentation, aiming to advance the field.
%R 10.18653/v1/2025.findings-naacl.200
%U https://aclanthology.org/2025.findings-naacl.200/
%U https://doi.org/10.18653/v1/2025.findings-naacl.200
%P 3621-3634
Markdown (Informal)
[When and How to Augment Your Input: Question Routing Helps Balance the Accuracy and Efficiency of Large Language Models](https://aclanthology.org/2025.findings-naacl.200/) (Chen et al., Findings 2025)
ACL