Reimagining Peer Review Process Through Multi-Agent Mechanism Design
This addresses the problem of failing peer review processes for the software engineering research community, but it is a vision paper outlining a research agenda rather than presenting results, so it is incremental in nature.
The paper tackles the systemic crisis of peer review in software engineering research by proposing to model the community as a stochastic multi-agent system and apply multi-agent reinforcement learning to design incentive-compatible protocols, outlining interventions like a credit-based submission economy and MARL-optimized reviewer assignment.
The software engineering research community faces a systemic crisis: peer review is failing under growing submissions, misaligned incentives, and reviewer fatigue. Community surveys reveal that researchers perceive the process as "broken." This position paper argues that these dysfunctions are mechanism design failures amenable to computational solutions. We propose modeling the research community as a stochastic multi-agent system and applying multi-agent reinforcement learning to design incentive-compatible protocols. We outline three interventions: a credit-based submission economy, MARL-optimized reviewer assignment, and hybrid verification of review consistency. We present threat models, equity considerations, and phased pilot metrics. This vision charts a research agenda toward sustainable peer review.