Post-Training Fairness Control: A Single-Train Framework for Dynamic Fairness in Recommendation

TL;DR

Cofair is a novel framework that allows post-training adjustment of fairness levels in recommendation systems, enabling dynamic fairness control without retraining, while maintaining or improving accuracy.

Contribution

We introduce Cofair, a single-train method with fairness-conditioned adapters and user-level regularization for flexible, post-training fairness adjustment in recommendation systems.

Findings

Provides dynamic fairness control at different levels

Achieves comparable or better fairness-accuracy trade-offs

Eliminates the need for retraining for new fairness requirements

Abstract

Despite growing efforts to mitigate unfairness in recommender systems, existing fairness-aware methods typically fix the fairness requirement at training time and provide limited post-training flexibility. However, in real-world scenarios, diverse stakeholders may demand differing fairness requirements over time, so retraining for different fairness requirements becomes prohibitive. To address this limitation, we propose Cofair, a single-train framework that enables post-training fairness control in recommendation. Specifically, Cofair introduces a shared representation layer with fairness-conditioned adapter modules to produce user embeddings specialized for varied fairness levels, along with a user-level regularization term that guarantees user-wise monotonic fairness improvements across these levels. We theoretically establish that the adversarial objective of Cofair upper bounds demographic parity and the regularization term enforces progressive fairness at user level. Comprehensive experiments on multiple datasets and backbone models demonstrate that our framework provides dynamic fairness at different levels, delivering comparable or better fairness-accuracy curves than state-of-the-art baselines, without the need to retrain for each new fairness requirement. Our code is publicly available at https://github.com/weixinchen98/Cofair.