@inproceedings{al-ali-etal-2026-green,
title = "Green Bots versus Red Bots: Evaluating Large Language Models for Simulating Persuasion Dynamics in Online Influence Campaigns",
author = "Al Ali, Majd Eddin and
Muntean, Filip Mihai and
Donatelli, Lucia and
van Diggelen, Jurriaan",
editor = "Piperidis, Stelios and
Bel, N{\'u}ria and
van den Heuvel, Henk and
Ide, Nancy and
Krek, Simon and
Toral, Antonio",
booktitle = "Proceedings of the Fifteenth Language Resources and Evaluation Conference",
month = may,
year = "2026",
address = "Palma de Mallorca, Spain",
publisher = "ELRA Language Resource Association",
url = "https://aclanthology.org/2026.lrec-1.871/",
doi = "10.63317/3ucmt3dnin5n",
pages = "11152--11171",
abstract = "Large language models (LLMs) are increasingly used to simulate social interaction and persuasion dynamics, yet their validity as proxies for human cognition and behavior remains unverified. We propose a dual-level evaluation framework to assess LLM-based agents at both the individual and collective levels. At the individual level, we examine agent fidelity by comparing LLM-generated political personas to human benchmark data. We find that while agents capture broad partisan orientations, they underestimate within-group variability and reproduce stereotypical ideological biases. At the collective level, we deploy Big Five personality-differentiated agents in 1080 structured dialogues to test the effect of rhetorical strategy on persuasive success. Our simulations reproduce theoretically expected interaction patterns; nevertheless, belief shifts are exaggerated relative to human baselines, supporting LLMs' tendency toward over-responsiveness. These findings suggest a trade-off between engagement-optimized training objectives and psychological realism, confirming the need to use LLMs with caution to simulate human behavior. We contribute three resources: a persuasion dynamics dataset, a standardized agent taxonomy of ``red'' and ``green'' bots, and a framework for evaluating both individual-agent fidelity and emergent group-level behavior."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="al-ali-etal-2026-green">
<titleInfo>
<title>Green Bots versus Red Bots: Evaluating Large Language Models for Simulating Persuasion Dynamics in Online Influence Campaigns</title>
</titleInfo>
<name type="personal">
<namePart type="given">Majd</namePart>
<namePart type="given">Eddin</namePart>
<namePart type="family">Al Ali</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Filip</namePart>
<namePart type="given">Mihai</namePart>
<namePart type="family">Muntean</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lucia</namePart>
<namePart type="family">Donatelli</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jurriaan</namePart>
<namePart type="family">van Diggelen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Fifteenth Language Resources and Evaluation Conference</title>
</titleInfo>
<name type="personal">
<namePart type="given">Stelios</namePart>
<namePart type="family">Piperidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Núria</namePart>
<namePart type="family">Bel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Henk</namePart>
<namePart type="family">van den Heuvel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nancy</namePart>
<namePart type="family">Ide</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Simon</namePart>
<namePart type="family">Krek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antonio</namePart>
<namePart type="family">Toral</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resource Association</publisher>
<place>
<placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Large language models (LLMs) are increasingly used to simulate social interaction and persuasion dynamics, yet their validity as proxies for human cognition and behavior remains unverified. We propose a dual-level evaluation framework to assess LLM-based agents at both the individual and collective levels. At the individual level, we examine agent fidelity by comparing LLM-generated political personas to human benchmark data. We find that while agents capture broad partisan orientations, they underestimate within-group variability and reproduce stereotypical ideological biases. At the collective level, we deploy Big Five personality-differentiated agents in 1080 structured dialogues to test the effect of rhetorical strategy on persuasive success. Our simulations reproduce theoretically expected interaction patterns; nevertheless, belief shifts are exaggerated relative to human baselines, supporting LLMs’ tendency toward over-responsiveness. These findings suggest a trade-off between engagement-optimized training objectives and psychological realism, confirming the need to use LLMs with caution to simulate human behavior. We contribute three resources: a persuasion dynamics dataset, a standardized agent taxonomy of “red” and “green” bots, and a framework for evaluating both individual-agent fidelity and emergent group-level behavior.</abstract>
<identifier type="citekey">al-ali-etal-2026-green</identifier>
<identifier type="doi">10.63317/3ucmt3dnin5n</identifier>
<location>
<url>https://aclanthology.org/2026.lrec-1.871/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>11152</start>
<end>11171</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Green Bots versus Red Bots: Evaluating Large Language Models for Simulating Persuasion Dynamics in Online Influence Campaigns
%A Al Ali, Majd Eddin
%A Muntean, Filip Mihai
%A Donatelli, Lucia
%A van Diggelen, Jurriaan
%Y Piperidis, Stelios
%Y Bel, Núria
%Y van den Heuvel, Henk
%Y Ide, Nancy
%Y Krek, Simon
%Y Toral, Antonio
%S Proceedings of the Fifteenth Language Resources and Evaluation Conference
%D 2026
%8 May
%I ELRA Language Resource Association
%C Palma de Mallorca, Spain
%F al-ali-etal-2026-green
%X Large language models (LLMs) are increasingly used to simulate social interaction and persuasion dynamics, yet their validity as proxies for human cognition and behavior remains unverified. We propose a dual-level evaluation framework to assess LLM-based agents at both the individual and collective levels. At the individual level, we examine agent fidelity by comparing LLM-generated political personas to human benchmark data. We find that while agents capture broad partisan orientations, they underestimate within-group variability and reproduce stereotypical ideological biases. At the collective level, we deploy Big Five personality-differentiated agents in 1080 structured dialogues to test the effect of rhetorical strategy on persuasive success. Our simulations reproduce theoretically expected interaction patterns; nevertheless, belief shifts are exaggerated relative to human baselines, supporting LLMs’ tendency toward over-responsiveness. These findings suggest a trade-off between engagement-optimized training objectives and psychological realism, confirming the need to use LLMs with caution to simulate human behavior. We contribute three resources: a persuasion dynamics dataset, a standardized agent taxonomy of “red” and “green” bots, and a framework for evaluating both individual-agent fidelity and emergent group-level behavior.
%R 10.63317/3ucmt3dnin5n
%U https://aclanthology.org/2026.lrec-1.871/
%U https://doi.org/10.63317/3ucmt3dnin5n
%P 11152-11171
Markdown (Informal)
[Green Bots versus Red Bots: Evaluating Large Language Models for Simulating Persuasion Dynamics in Online Influence Campaigns](https://aclanthology.org/2026.lrec-1.871/) (Al Ali et al., LREC 2026)
ACL