The Illusion of Control: Commensurability, Concentration, and Correlation in Multi-Objective Recommendation

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

Multi-objective recommender systems (MORS) often combine diverse goals via linear scalarization, where objectives’ weights are being exposed as tunable hyperparameters for system optimization, or as interactive knobs supporting user control. However, we argue that such an approach creates an illusion of control, in which some objectives mechanically dominate the recommendations, and standard normalization techniques do not alleviate the issue. We formalize this through the 3C (Commensurability, Concentration, and Correlation) Bias Framework and introduce the Objective Dominance Ratio (ODR) to diagnose it. We benchmark four widely used normalization schemes on several datasets, identify its failure modes, and propose Rank-ZCA normalization, which is–to the best of our knowledge–the only strategy resilient against all 3C biases. Source codes & raw results are available from https://osf.io/z8sb2/.