@inproceedings{pan-etal-2026-coopagent,
title = "{C}oop{A}gent: Training Multi-Agent Cooperation from Single-Agent Execution Trajectories",
author = "Pan, Zhengzhe and
Zhu, Linchao and
Yang, Yi",
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.892/",
doi = "10.18653/2026.findings-emnlp.892",
pages = "18891--18909",
ISBN = "979-8-89176-462-0",
abstract = "LLM-based multi-agent systems have demonstrated strong potential for solving complex tasks, yet reliable collaboration remains challenging. While most training-based methods rely on multi-agent trajectories, we investigate whether cooperative behaviors can be strengthened using only single-agent execution data. We propose CoopAgent, a training framework that learns solely from single-agent execution trajectories and transfers to multi-agent deployment. To bridge the gap between single-agent training and multi-agent environments, we introduce PADPO (Perturbation-Aligned DPO), which aligns action preferences between original observations and task-related perturbations that mimic changes caused by other agents. We train CoopAgent on single-agent execution trajectories collected in Minecraft and evaluate it on VillagerBench. The results demonstrate its potential to enhance collaboration on VillagerBench while preserving strong single-agent execution ability. We further assess CoopAgent on out-of-domain benchmarks, showing improvements in several multi-agent settings and performance comparable to the multi-agent training method AT-GRPO on the coding benchmarks. Our code is available at https://github.com/hlpzz/CoopAgent."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="pan-etal-2026-coopagent">
<titleInfo>
<title>CoopAgent: Training Multi-Agent Cooperation from Single-Agent Execution Trajectories</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhengzhe</namePart>
<namePart type="family">Pan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Linchao</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yi</namePart>
<namePart type="family">Yang</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>LLM-based multi-agent systems have demonstrated strong potential for solving complex tasks, yet reliable collaboration remains challenging. While most training-based methods rely on multi-agent trajectories, we investigate whether cooperative behaviors can be strengthened using only single-agent execution data. We propose CoopAgent, a training framework that learns solely from single-agent execution trajectories and transfers to multi-agent deployment. To bridge the gap between single-agent training and multi-agent environments, we introduce PADPO (Perturbation-Aligned DPO), which aligns action preferences between original observations and task-related perturbations that mimic changes caused by other agents. We train CoopAgent on single-agent execution trajectories collected in Minecraft and evaluate it on VillagerBench. The results demonstrate its potential to enhance collaboration on VillagerBench while preserving strong single-agent execution ability. We further assess CoopAgent on out-of-domain benchmarks, showing improvements in several multi-agent settings and performance comparable to the multi-agent training method AT-GRPO on the coding benchmarks. Our code is available at https://github.com/hlpzz/CoopAgent.</abstract>
<identifier type="citekey">pan-etal-2026-coopagent</identifier>
<identifier type="doi">10.18653/2026.findings-emnlp.892</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.892/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>18891</start>
<end>18909</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CoopAgent: Training Multi-Agent Cooperation from Single-Agent Execution Trajectories
%A Pan, Zhengzhe
%A Zhu, Linchao
%A Yang, Yi
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%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 pan-etal-2026-coopagent
%X LLM-based multi-agent systems have demonstrated strong potential for solving complex tasks, yet reliable collaboration remains challenging. While most training-based methods rely on multi-agent trajectories, we investigate whether cooperative behaviors can be strengthened using only single-agent execution data. We propose CoopAgent, a training framework that learns solely from single-agent execution trajectories and transfers to multi-agent deployment. To bridge the gap between single-agent training and multi-agent environments, we introduce PADPO (Perturbation-Aligned DPO), which aligns action preferences between original observations and task-related perturbations that mimic changes caused by other agents. We train CoopAgent on single-agent execution trajectories collected in Minecraft and evaluate it on VillagerBench. The results demonstrate its potential to enhance collaboration on VillagerBench while preserving strong single-agent execution ability. We further assess CoopAgent on out-of-domain benchmarks, showing improvements in several multi-agent settings and performance comparable to the multi-agent training method AT-GRPO on the coding benchmarks. Our code is available at https://github.com/hlpzz/CoopAgent.
%R 10.18653/2026.findings-emnlp.892
%U https://aclanthology.org/2026.findings-emnlp.892/
%U https://doi.org/10.18653/2026.findings-emnlp.892
%P 18891-18909
Markdown (Informal)
[CoopAgent: Training Multi-Agent Cooperation from Single-Agent Execution Trajectories](https://aclanthology.org/2026.findings-emnlp.892/) (Pan et al., Findings 2026)
ACL