aiqbench analysis
This research introduces a multi-agent simulacra framework using LLM-driven agents to model and detect collusive bidding rings within the peer review lifecycle. By simulating honest versus collusive policies, the study quantifies how such coordination inflates target paper capture and review scores while revealing limitations in current bid-phase detectors. The findings offer critical insights into systemic integrity risks for major AI conferences and the need for more robust detection mechanisms against expertise-consistent attacks.
CABAL: Multi-Agent Simulacra for Tracing the Effects of Collusive Bidding in Peer Review
arXiv:2609.05227v1 Announce Type: new Abstract: Recent reports during the AAAI-27 review cycle highlight the risk of reviewers coordinating bids for reciprocal assignment advantage. Prior work treats bidding, reviewer assignment, and review manipulation as separate stages, leaving the lifecycle effects of collusive bidding unclear. Real-world analysis is further constrained by typically unobservable collusive intent and the lack of counterfactuals for the same conference. Motivated by this gap, we introduce \alg, 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. We further develop an affinity-guided collusive bidding strategy that uses mutual reviewer-paper affinities to construct collusion rings and select target papers, producing expertise-consistent rather than arbitrarily targeted attacks. Controlled experiments show that collusive bidding more than doubles target-paper capture and that assigned colluders score target papers about two points higher than honest co-reviewers, while conference-wide effects remain comparatively modest. Evaluated bid-phase detectors provide only limited evidence of collusion: in a fixed-triplet detector stress test, native positive-bid graphs are confounded by benign affinity, while a Very-High-only diagnostic view enables precise but low-coverage local recovery.
Abstract:Recent reports during the AAAI-27 review cycle highlight the risk of reviewers coordinating bids for reciprocal assignment advantage. Prior work treats bidding, reviewer assignment, and review manipulation as separate stages, leaving the lifecycle effects of collusive bidding unclear. Real-world analysis is further constrained by typically unobservable collusive intent and the lack of counterfactuals for the same conference. Motivated by this gap, we introduce \alg, 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. We further develop an affinity-guided collusive bidding strategy that uses mutual reviewer-paper affinities to construct collusion rings and select target papers, producing expertise-consistent rather than arbitrarily targeted attacks. Controlled experiments show that collusive bidding more than doubles target-paper capture and that assigned colluders score target papers about two points higher than honest co-reviewers, while conference-wide effects remain comparatively modest. Evaluated bid-phase detectors provide only limited evidence of collusion: in a fixed-triplet detector stress test, native positive-bid graphs are confounded by benign affinity, while a Very-High-only diagnostic view enables precise but low-coverage local recovery.
Artificial Intelligence (cs.AI)
arXiv:2609.05227 [cs.AI]
(or arXiv:2609.05227v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.05227 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 277 words
Related coverage
Game-Theoretic Disempowerment: Loss of Control to Agencyless AI
LessWrong · Sep 27, 2026
State of the Field: AI for Epistemics and Coordination
LessWrong · Oct 2, 2026
Beyond the Transcript: Detecting Covert Co ordination in Latent Multi-Agent Communication
arXiv · Aug 21, 2026
Debating RSI, the US-China Gap, and Jaggedness with JS Denain of Epoch AI
Interconnects · Sep 22, 2026