A Unified Framework for Scalable and Robust Paper Assignment

2026-01-20

AI Summary

This work addresses the challenge of assigning a large volume of academic submissions to reviewers while simultaneously optimizing for expertise matching, assignment diversity, and robustness against strategic manipulation—a multi-objective problem that existing methods struggle to balance under scalability constraints. The authors propose RAMP, a novel framework that unifies these three objectives into a single optimization model for the first time. By leveraging linearized worst-case perturbation maximization and soft constraints, RAMP enables efficient computation, and an attribute-aware sampling strategy converts fractional solutions into integral assignments. Evaluated on a real-world dataset comprising over 20,000 papers and 20,000 reviewers, RAMP completes assignment in under 20 minutes, significantly outperforming state-of-the-art approaches in both scalability and overall performance.

Technology Category

• Machine Learning: Learning Preferences or Rankings
• Search and Optimization: Metareasoning and Metaheuristics
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Assigning papers to reviewers is a central challenge in the peer-review process of large academic conferences. Program chairs must balance competing objectives, including maximizing reviewer expertise, promoting diversity, and enhancing robustness to strategic manipulation, but it is challenging to do so at the modern conference scale. Existing algorithmic paper assignment approaches either fail to address all of these goals simultaneously or suffer from poor scalability. To address the limitation, we propose Robust Assignment via Marginal Perturbation (RAMP), a unified framework for large-scale peer review. Our approach formulates a linearized perturbed-maximization objective with soft constraints that flexibly balance assignment quality, diversity, and robustness while maintaining runtime efficiency. We further introduce an attribute-aware sampling procedure that converts fractional solutions into integral assignments and improves the diversity and robustness of the final assignment. On datasets with over 20,000 papers and 20,000 reviewers, RAMP runs in under 20 minutes, demonstrating its suitability for real-world deployment.

Problem

• paper assignment
• peer review
• scalability
• robustness
• diversity

Innovation

• paper assignment
• robustness
• diversity
• scalability
• marginal perturbation