Will the Saboteur Win? Robust Coordination Under Adversarial Behaviour in LLM Agent Teams

Institut

Professur für autonome Fahrzeugsysteme (TUM-ED)

Typ

Masterarbeit /

Inhalt

 

Beschreibung

Multi-agent LLM systems promise to solve complex tasks through collaboration. Yet their performance may depend heavily on whether a coordinating agent can recognise reliable, unreliable, or strategically misleading contributions over time. This thesis explores how an orchestrator can use interaction history to make robust coordination decisions under imperfect or adversarial agent behaviour.

Key facts:
Type: MA, also for Informatics students
Starting Date: Flexible
Supervisor: Prof. Dr.-Ing. Johannes Betz
Advisor: Yuchen Zhang, M.Sc
Preferred Experience: Python, LLM

If you are interested, simply send an email with your CV and academic transcript to yuchen2.zhang@tum.de.

Tags

AVS Zhang

Möglicher Beginn

sofort

Kontakt

Yuchen Zhang
yuchen2.zhangtum.de