paper-with-me

Papers

The Subtle Art of Defection: Understanding Uncooperative Behaviors in LLM based Multi-Agent Systems

2025-11-19 · Devang Kulshreshtha, Wanyu Du, Raghav Jain, Srikanth Doss, Hang Su, Sandesh Swamy, Yanjun Qi arxiv

This paper introduces a novel framework for simulating and analyzing how uncooperative behaviors can destabilize or collapse LLM-based multi-agent systems. Our framework includes two key components: (1) a game theory-based taxonomy of uncooperative agent behaviors, addressing a notable gap in the existing literature; and (2) a structured, multi-stage simulation pipeline that dynamically generates and refines uncooperative behaviors as agents' states evolve. We evaluate the framework via a collaborative resource management setting, measuring system stability using metrics such as survival time and resource overuse rate. Empirically, our framework achieves 96.7% accuracy in generating realistic uncooperative behaviors, validated by human evaluations. Our results reveal a striking contrast: cooperative agents maintain perfect system stability (100% survival over 12 rounds with 0% resource overuse), while any uncooperative behavior can trigger rapid system collapse within 1 to 7 rounds. We also evaluate LLM-based defense methods, finding they detect some uncooperative behaviors, but some behaviors remain largely undetectable. These gaps highlight how uncooperative agents degrade collective outcomes and underscore the need for more resilient multi-agent systems.

📄 PDF Abstract BibTeX arXiv:2511.15862

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Nicer Than Humans: How do Large Language Models Behave in the Prisoner's Dilemma?

2024-06-19 · Nicoló Fontana, Francesco Pierri, Luca Maria Aiello

The behavior of Large Language Models (LLMs) as artificial social agents is largely unexplored, and we still lack extensive evidence of how these agents react to simple social stimuli. Testing the behavior of AI agents i…

Decision Making

Peer-Preservation in Frontier Models

2026-03-30 · Yujin Potter, Nicholas Crispino, Vincent Siu, Chenguang Wang 외 arxiv

Recent work has found that frontier AI models can exhibit misaligned behaviors in pursuit of assigned goals. We demonstrate that models can also exhibit misaligned behaviors in defiance of assigned goals, appearing to se…

Coevolutionary dynamics of feedback-evolving games in structured populations

2025-02-09 · Qiushuang Wang, Xiaojie Chen, Attila Szolnoki

The interdependence between an individual strategy decision and the resulting change of environmental state is often a subtle process. Feedback-evolving games have been a prevalent framework for studying such feedback in…

Evolution of cooperation in a three-strategy game combining snowdrift and stag hunt games

2024-11-08 · Hirofumi Takesue

This study aimed to investigate the evolutionary dynamics of a three-strategy game that combines snowdrift and stag hunt games. This game is motivated by an experimental study, which found that individual solution lowers…

Stereo Event-based, 6-DOF Pose Tracking for Uncooperative Spacecraft

2025-03-17 · Zibin Liu, Banglei Guan, Yang Shang, Yifei Bian 외

Pose tracking of uncooperative spacecraft is an essential technology for space exploration and on-orbit servicing, which remains an open problem. Event cameras possess numerous advantages, such as high dynamic range, hig…

Pose Tracking