CABAL: Multi-Agent Simulacra for Tracing the Effects of Collusive Bidding in Peer Review
Alg is introduced, an end-to-end multi-agent simulacra framework for studying reviewer assignment integrity by holding the conference environment fixed and configuring LLM-driven reviewer agents with honest or collusive policies, and an affinity-guided collusive bidding strategy that uses mutual reviewer-paper affiniti...
Ji-Cheng Zhou, Ke-Mou Li, Kahim Wong et al.
· 1 citation