@inproceedings{zou-etal-2026-ai,
title = "From {AI} Scientists to {AI} Scientific Communities: A Survey of Collective {A}uto{R}esearch",
author = "Zou, Rui and
Xu, Tianyi and
Zhang, Fan and
Zhang, Yao and
Zhao, Zhe and
Xiaolongwang and
Wang, Maolin and
Wu, Yingcheng and
Wei, Ye",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.824/",
pages = "17458--17472",
ISBN = "979-8-89176-462-0",
abstract = "Scientific research is not merely a sequence of tasks to be automated, but a knowledge-production process in which claims must remain verifiable, criticizable, revisable, and accountable. Yet most AutoResearch systems remain single scaffolds that collapse generation, execution, interpretation, and review into one loop. This survey diagnoses three resulting epistemic limits of such architectures: collapsed skepticism, where a single scaffold produces not only errors but also the narratives that allow them to survive; unverified translation from discovery to application, where proposal generation outpaces independent verification; and invisible evidence trails, where final manuscripts hide the failed attempts and post-hoc decisions that produced them. To overcome these structural fragilities, we propose a shift from autonomous AI scientists to AI research communities, where specialized agents, shared memory, inspectable artifacts, and governance checkpoints distribute epistemic labor. We synthesize AutoResearch as a transition from metric-driven search loops to AI-scientist workflows, domain-mediated discovery, research infrastructure, and accountable human{--}AI co-evolution. Across these layers, we argue for a fundamental shift from productivity-centered evaluation to verifiability-centered evaluation."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zou-etal-2026-ai">
<titleInfo>
<title>From AI Scientists to AI Scientific Communities: A Survey of Collective AutoResearch</title>
</titleInfo>
<name type="personal">
<namePart type="given">Rui</namePart>
<namePart type="family">Zou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tianyi</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fan</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yao</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhe</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name>
<namePart>Xiaolongwang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maolin</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yingcheng</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ye</namePart>
<namePart type="family">Wei</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>Findings of the Association for Computational Linguistics: EMNLP 2026</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">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Scientific research is not merely a sequence of tasks to be automated, but a knowledge-production process in which claims must remain verifiable, criticizable, revisable, and accountable. Yet most AutoResearch systems remain single scaffolds that collapse generation, execution, interpretation, and review into one loop. This survey diagnoses three resulting epistemic limits of such architectures: collapsed skepticism, where a single scaffold produces not only errors but also the narratives that allow them to survive; unverified translation from discovery to application, where proposal generation outpaces independent verification; and invisible evidence trails, where final manuscripts hide the failed attempts and post-hoc decisions that produced them. To overcome these structural fragilities, we propose a shift from autonomous AI scientists to AI research communities, where specialized agents, shared memory, inspectable artifacts, and governance checkpoints distribute epistemic labor. We synthesize AutoResearch as a transition from metric-driven search loops to AI-scientist workflows, domain-mediated discovery, research infrastructure, and accountable human–AI co-evolution. Across these layers, we argue for a fundamental shift from productivity-centered evaluation to verifiability-centered evaluation.</abstract>
<identifier type="citekey">zou-etal-2026-ai</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.824/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>17458</start>
<end>17472</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T From AI Scientists to AI Scientific Communities: A Survey of Collective AutoResearch
%A Zou, Rui
%A Xu, Tianyi
%A Zhang, Fan
%A Zhang, Yao
%A Zhao, Zhe
%A Wang, Maolin
%A Wu, Yingcheng
%A Wei, Ye
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%A Xiaolongwang
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F zou-etal-2026-ai
%X Scientific research is not merely a sequence of tasks to be automated, but a knowledge-production process in which claims must remain verifiable, criticizable, revisable, and accountable. Yet most AutoResearch systems remain single scaffolds that collapse generation, execution, interpretation, and review into one loop. This survey diagnoses three resulting epistemic limits of such architectures: collapsed skepticism, where a single scaffold produces not only errors but also the narratives that allow them to survive; unverified translation from discovery to application, where proposal generation outpaces independent verification; and invisible evidence trails, where final manuscripts hide the failed attempts and post-hoc decisions that produced them. To overcome these structural fragilities, we propose a shift from autonomous AI scientists to AI research communities, where specialized agents, shared memory, inspectable artifacts, and governance checkpoints distribute epistemic labor. We synthesize AutoResearch as a transition from metric-driven search loops to AI-scientist workflows, domain-mediated discovery, research infrastructure, and accountable human–AI co-evolution. Across these layers, we argue for a fundamental shift from productivity-centered evaluation to verifiability-centered evaluation.
%U https://aclanthology.org/2026.findings-emnlp.824/
%P 17458-17472
Markdown (Informal)
[From AI Scientists to AI Scientific Communities: A Survey of Collective AutoResearch](https://aclanthology.org/2026.findings-emnlp.824/) (Zou et al., Findings 2026)
ACL
- Rui Zou, Tianyi Xu, Fan Zhang, Yao Zhang, Zhe Zhao, Xiaolongwang, Maolin Wang, Yingcheng Wu, and Ye Wei. 2026. From AI Scientists to AI Scientific Communities: A Survey of Collective AutoResearch. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 17458–17472, Budapest, Hungary. Association for Computational Linguistics.