发表机构
University of Macau; Mila – Québec AI Institute; McGill University; King Abdullah University of Science and Technology; University of Electronic Science and Technology of China（澳门大学; 米拉-魁北克人工智能研究所; 麦吉尔大学; 阿卜杜拉国王科技大学; 电子科技大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对同行评审中合谋投标的问题，提出多智能体模拟框架CABAL，开发亲和度引导的合谋投标策略，经实验验证其对目标论文捕获的影响及检测器的局限。
AI 中文摘要
近期AAAI-27评审周期的报告强调了评审人员协调投标以获取互惠分配优势的风险。现有研究将投标、评审人员分配和评审操纵视为独立阶段，导致合谋投标的生命周期效应尚不明确。实际分析还受到合谋意图通常不可观测以及同一会议缺乏反事实数据的限制。针对这一缺口，我们引入CABAL，这是一个端到端的多智能体模拟框架，用于研究评审人员分配的完整性，方法是固定会议环境并配置由大语言模型（LLM）驱动的、具有诚实或合谋策略的评审人员智能体。我们进一步开发了一种亲和度引导的合谋投标策略，该策略利用评审人员与论文之间的相互亲和度来构建合谋圈子并选择目标论文，从而产生与专业知识一致而非任意针对性的攻击。受控实验表明，合谋投标使目标论文的捕获量增加了一倍以上，且被分配的合谋者对目标论文的评分比诚实的共同评审人员高出约2分，而会议范围内的影响相对较小。对投标阶段检测器的评估仅提供有限的合谋证据：在固定三元组检测器的压力测试中，原生正投标图被良性亲和度混淆，而仅针对极高值的诊断视图能够实现精确但覆盖范围低的局部恢复。
英文摘要
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.
