@inproceedings{guo-etal-2026-surveylens,
title = "{S}urvey{L}ens: A Discipline-Aware Benchmark for Automatic Survey Generation",
author = "Guo, Beichen and
Wen, Zhiyuan and
GU, Jia and
Shi, Haochen and
Wang, Jian and
Wang, Senzhang and
LI, Haoyang and
Yang, Ruosong and
Liu, Shuaiqi",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.2616/",
pages = "59426--59458",
ISBN = "979-8-89176-459-0",
abstract = "Automatic Survey Generation (ASG) aims to produce comprehensive literature surveys by retrieving, organizing, and synthesizing academic papers. Despite rapid progress in specialized ASG frameworks and Deep Research agents, existing evaluations largely center on Computer Science or rely on generic criteria, leaving it unclear whether current systems satisfy the survey standards of diverse disciplines. We introduce SurveyLens, the first discipline-aware ASG benchmark. SurveyLens comprises SurveyLens-1k, a curated dataset of 1,000 human-written surveys across 10 disciplines, and a dual-lens framework that combines discipline-aware rubric scoring with reference-based alignment to human-written surveys. Evaluating 11 state-of-the-art systems spanning vanilla LLMs, specialized ASG systems, and Deep Research agents, we find that Deep Research agents perform most consistently across the 10 disciplines, some specialized ASG systems show particular strength in outline generation, and reference coverage and relevance remain a shared bottleneck across paradigms. These findings further provide practical guidance for discipline-specific system selection and future ASG design."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="guo-etal-2026-surveylens">
<titleInfo>
<title>SurveyLens: A Discipline-Aware Benchmark for Automatic Survey Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Beichen</namePart>
<namePart type="family">Guo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhiyuan</namePart>
<namePart type="family">Wen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jia</namePart>
<namePart type="family">GU</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haochen</namePart>
<namePart type="family">Shi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jian</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Senzhang</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haoyang</namePart>
<namePart type="family">LI</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruosong</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shuaiqi</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Automatic Survey Generation (ASG) aims to produce comprehensive literature surveys by retrieving, organizing, and synthesizing academic papers. Despite rapid progress in specialized ASG frameworks and Deep Research agents, existing evaluations largely center on Computer Science or rely on generic criteria, leaving it unclear whether current systems satisfy the survey standards of diverse disciplines. We introduce SurveyLens, the first discipline-aware ASG benchmark. SurveyLens comprises SurveyLens-1k, a curated dataset of 1,000 human-written surveys across 10 disciplines, and a dual-lens framework that combines discipline-aware rubric scoring with reference-based alignment to human-written surveys. Evaluating 11 state-of-the-art systems spanning vanilla LLMs, specialized ASG systems, and Deep Research agents, we find that Deep Research agents perform most consistently across the 10 disciplines, some specialized ASG systems show particular strength in outline generation, and reference coverage and relevance remain a shared bottleneck across paradigms. These findings further provide practical guidance for discipline-specific system selection and future ASG design.</abstract>
<identifier type="citekey">guo-etal-2026-surveylens</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.2616/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>59426</start>
<end>59458</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SurveyLens: A Discipline-Aware Benchmark for Automatic Survey Generation
%A Guo, Beichen
%A Wen, Zhiyuan
%A GU, Jia
%A Shi, Haochen
%A Wang, Jian
%A Wang, Senzhang
%A LI, Haoyang
%A Yang, Ruosong
%A Liu, Shuaiqi
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F guo-etal-2026-surveylens
%X Automatic Survey Generation (ASG) aims to produce comprehensive literature surveys by retrieving, organizing, and synthesizing academic papers. Despite rapid progress in specialized ASG frameworks and Deep Research agents, existing evaluations largely center on Computer Science or rely on generic criteria, leaving it unclear whether current systems satisfy the survey standards of diverse disciplines. We introduce SurveyLens, the first discipline-aware ASG benchmark. SurveyLens comprises SurveyLens-1k, a curated dataset of 1,000 human-written surveys across 10 disciplines, and a dual-lens framework that combines discipline-aware rubric scoring with reference-based alignment to human-written surveys. Evaluating 11 state-of-the-art systems spanning vanilla LLMs, specialized ASG systems, and Deep Research agents, we find that Deep Research agents perform most consistently across the 10 disciplines, some specialized ASG systems show particular strength in outline generation, and reference coverage and relevance remain a shared bottleneck across paradigms. These findings further provide practical guidance for discipline-specific system selection and future ASG design.
%U https://aclanthology.org/2026.emnlp-main.2616/
%P 59426-59458
Markdown (Informal)
[SurveyLens: A Discipline-Aware Benchmark for Automatic Survey Generation](https://aclanthology.org/2026.emnlp-main.2616/) (Guo et al., EMNLP 2026)
ACL
- Beichen Guo, Zhiyuan Wen, Jia GU, Haochen Shi, Jian Wang, Senzhang Wang, Haoyang LI, Ruosong Yang, and Shuaiqi Liu. 2026. SurveyLens: A Discipline-Aware Benchmark for Automatic Survey Generation. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 59426–59458, Budapest, Hungary. Association for Computational Linguistics.