• Classified by Topic • Classified by Publication Type • Sorted by Date • Sorted by First Author Last Name • Classified by Funding Source •
CoopReflect: Towards Natural Language Communication for Cooperative Autonomous Driving via Multi-Agent Learning.
Jiaxun Cui, Chen Tang, Jarrett Holtz,
Janice Nguyen, Alessandro G Allievi, Hang Qiu, and Peter Stone.
In International
Conference on Autonomous Agents and Multi-Agent Systems, May 2026.
Past work has demonstrated that autonomous vehicles can drive more safely if theycommunicate with each other. However, this communication is usually nothuman-understandable. Using natural language as a vehicle-to-vehicle (V2V)communication protocol offers the potential for autonomous vehicles to drivecooperatively not only with each other but also with human drivers. To explorethe potential use of natural language for V2V communication, we develop LLM-baseddriving agents and study their interactions in a new simulation environment,TalkingVehiclesGym, which features traffic scenarios where communication canpotentially help avoid imminent collisions and/or support efficient traffic flow.While LLM agents relying solely on chain-of-thought reasoning struggle tocoordinate effectively, we introduce CoopReflect, a multi-agent learningframework that equips agents with knowledge for both natural language messagegeneration and high-level decision-making through trial and error and multi-agentdebriefing. Experiments show that CoopReflect produces more meaningful andhuman-understandable messages than existing baselines, enabling strongercooperation. Finally, we distill scenario-specific knowledge into a unifiedlanguage model policy, achieving cross-scenario generalization and substantiallyreducing decision-making latency. Our code and demo videos are available athttps://talking-vehicles.github.io/.
@InProceedings{cui2026coopreflect,
author = {Jiaxun Cui and Chen Tang and Jarrett Holtz and Janice Nguyen and Alessandro G Allievi and Hang Qiu and Peter Stone},
title = {CoopReflect: Towards Natural Language Communication for Cooperative Autonomous Driving via Multi-Agent Learning},
booktitle = {International Conference on Autonomous Agents and Multi-Agent Systems},
year = {2026},
month = {May},
location = {Paphos, Cyprus},
abstract = {Past work has demonstrated that autonomous vehicles can drive more safely if they
communicate with each other. However, this communication is usually not
human-understandable. Using natural language as a vehicle-to-vehicle (V2V)
communication protocol offers the potential for autonomous vehicles to drive
cooperatively not only with each other but also with human drivers. To explore
the potential use of natural language for V2V communication, we develop LLM-based
driving agents and study their interactions in a new simulation environment,
TalkingVehiclesGym, which features traffic scenarios where communication can
potentially help avoid imminent collisions and/or support efficient traffic flow.
While LLM agents relying solely on chain-of-thought reasoning struggle to
coordinate effectively, we introduce CoopReflect, a multi-agent learning
framework that equips agents with knowledge for both natural language message
generation and high-level decision-making through trial and error and multi-agent
debriefing. Experiments show that CoopReflect produces more meaningful and
human-understandable messages than existing baselines, enabling stronger
cooperation. Finally, we distill scenario-specific knowledge into a unified
language model policy, achieving cross-scenario generalization and substantially
reducing decision-making latency. Our code and demo videos are available at
https://talking-vehicles.github.io/.
},
}
Generated by bib2html.pl (written by Patrick Riley ) on Mon Feb 23, 2026 19:28:57