@inproceedings{parfenova-etal-2025-emergent,
title = "Emergent Convergence in Multi-Agent {LLM} Annotation",
author = {Parfenova, Angelina and
Denzler, Alexander and
Pfeffer, J{\"u}rgen},
editor = "Belinkov, Yonatan and
Mueller, Aaron and
Kim, Najoung and
Mohebbi, Hosein and
Chen, Hanjie and
Arad, Dana and
Sarti, Gabriele",
booktitle = "Proceedings of the 8th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.blackboxnlp-1.12/",
pages = "206--225",
ISBN = "979-8-89176-346-3",
abstract = "Large language models (LLMs) are increasingly deployed in collaborative settings, yet little is known about how they coordinate when treated as black-box agents. We simulate 7,500 multi-agent, multi-round discussions in an inductive coding task, generating over 125,000 utterances that capture both final annotations and their interactional histories. We introduce process-level metrics{---}code stability, semantic self-consistency, and lexical confidence{---}alongside sentiment and convergence measures, to track coordination dynamics. To probe deeper alignment signals, we analyze the evolving geometry of output embeddings, showing that intrinsic dimensionality declines over rounds, suggesting semantic compression. The results reveal that LLM groups converge lexically and semantically, develop asymmetric influence patterns, and exhibit negotiation-like behaviors despite the absence of explicit role prompting. This work demonstrates how black-box interaction analysis can surface emergent coordination strategies, offering a scalable complement to internal probe-based interpretability methods."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="parfenova-etal-2025-emergent">
<titleInfo>
<title>Emergent Convergence in Multi-Agent LLM Annotation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Angelina</namePart>
<namePart type="family">Parfenova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexander</namePart>
<namePart type="family">Denzler</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jürgen</namePart>
<namePart type="family">Pfeffer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 8th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yonatan</namePart>
<namePart type="family">Belinkov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aaron</namePart>
<namePart type="family">Mueller</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Najoung</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hosein</namePart>
<namePart type="family">Mohebbi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hanjie</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dana</namePart>
<namePart type="family">Arad</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gabriele</namePart>
<namePart type="family">Sarti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-346-3</identifier>
</relatedItem>
<abstract>Large language models (LLMs) are increasingly deployed in collaborative settings, yet little is known about how they coordinate when treated as black-box agents. We simulate 7,500 multi-agent, multi-round discussions in an inductive coding task, generating over 125,000 utterances that capture both final annotations and their interactional histories. We introduce process-level metrics—code stability, semantic self-consistency, and lexical confidence—alongside sentiment and convergence measures, to track coordination dynamics. To probe deeper alignment signals, we analyze the evolving geometry of output embeddings, showing that intrinsic dimensionality declines over rounds, suggesting semantic compression. The results reveal that LLM groups converge lexically and semantically, develop asymmetric influence patterns, and exhibit negotiation-like behaviors despite the absence of explicit role prompting. This work demonstrates how black-box interaction analysis can surface emergent coordination strategies, offering a scalable complement to internal probe-based interpretability methods.</abstract>
<identifier type="citekey">parfenova-etal-2025-emergent</identifier>
<location>
<url>https://aclanthology.org/2025.blackboxnlp-1.12/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>206</start>
<end>225</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Emergent Convergence in Multi-Agent LLM Annotation
%A Parfenova, Angelina
%A Denzler, Alexander
%A Pfeffer, Jürgen
%Y Belinkov, Yonatan
%Y Mueller, Aaron
%Y Kim, Najoung
%Y Mohebbi, Hosein
%Y Chen, Hanjie
%Y Arad, Dana
%Y Sarti, Gabriele
%S Proceedings of the 8th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-346-3
%F parfenova-etal-2025-emergent
%X Large language models (LLMs) are increasingly deployed in collaborative settings, yet little is known about how they coordinate when treated as black-box agents. We simulate 7,500 multi-agent, multi-round discussions in an inductive coding task, generating over 125,000 utterances that capture both final annotations and their interactional histories. We introduce process-level metrics—code stability, semantic self-consistency, and lexical confidence—alongside sentiment and convergence measures, to track coordination dynamics. To probe deeper alignment signals, we analyze the evolving geometry of output embeddings, showing that intrinsic dimensionality declines over rounds, suggesting semantic compression. The results reveal that LLM groups converge lexically and semantically, develop asymmetric influence patterns, and exhibit negotiation-like behaviors despite the absence of explicit role prompting. This work demonstrates how black-box interaction analysis can surface emergent coordination strategies, offering a scalable complement to internal probe-based interpretability methods.
%U https://aclanthology.org/2025.blackboxnlp-1.12/
%P 206-225
Markdown (Informal)
[Emergent Convergence in Multi-Agent LLM Annotation](https://aclanthology.org/2025.blackboxnlp-1.12/) (Parfenova et al., BlackboxNLP 2025)
ACL
- Angelina Parfenova, Alexander Denzler, and Jürgen Pfeffer. 2025. Emergent Convergence in Multi-Agent LLM Annotation. In Proceedings of the 8th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP, pages 206–225, Suzhou, China. Association for Computational Linguistics.