Recommender systems align user rewards with their contributions
Aligning Performance with Contribution: Towards Contribution-Aware Fair Recommendation
Information Retrieval
Summary
Recommendation systems usually try to be fair but don’t often consider how much each user actually helps the system learn. The authors suggest that users who contribute more information should get better recommendations as a fair reward. They introduce a new method called CPFR that groups users by how much they contribute and tries to match their recommendation quality to their contribution. This approach can encourage users to stay engaged and improve overall system performance. Tests show CPFR balances accuracy and fairness well across different models and data.
What this means in practice
- •For online platform operators: Adjust recommendation strategies to reward users proportionally to their data contributions, boosting platform sustainability and user engagement.
- •For marketing analysts: Identify and target high-contributing user groups with tailored campaigns supported by contribution-aware recommendation metrics.
Authors
Shuai Zhang, Hui Fang, Zun Sun
Abstract
Existing research on user fairness in recommender systems has developed diverse objectives. However, it has paid limited attention to a distinct distributive perspective: whether users' contributions to model learning should be reflected in the recommendation benefits they receive. We argue that, in addition to existing fairness protections, a fair system may account for the alignment between users' estimated contributions and the recommendation performance they receive. Such alignment can incentivize sustained and informative engagement, thereby supporting a sustainable recommendation ecosystem. To this end, we propose Contribution-Performance Fairness, a novel fairness perspective which requires recommendation performance to be aligned with estimated contribution across user groups and to remain equitable among users with comparable contributions within a same group. To instantiate this perspective, we introduce the Contribution-Performance Fair Recommender (CPFR), a framework applicable to different backbone recommenders. CPFR constructs ordered user groups from a training-dependent contribution considering interaction volume, loss alignment, and optimization intensity, and jointly optimizes recommendation accuracy with the two fairness requirements. A game-theoretic analysis shows that such alignment can strengthen contribution incentives and improve system-level recommendation accuracy under voluntary contribution. Experiments on three datasets and three backbone models demonstrate that CPFR achieves a strong accuracy--fairness trade-off under the proposed operational metric.