Work

How Do Recommendation Models Amplify Popularity Bias? An Analysis from the Spectral Perspective

Year: 2025

Type: conference-paper

Abstract: Recommendation Systems (RS) are often plagued by popularity bias. When training a recommendation model on a typically long-tailed dataset, the model tends to not only inherit this bias but often exacerbate it, resulting in over-representation of popular items in the recommendation lists. This study conducts comprehensive empirical and theoretical analyses to expose the root causes of this phenomenon, yielding two core insights: 1) Item popularity is memorized in the principal spectrum of the score matrix predicted by the recommendation model; 2) The dimension reduction phenomenon amplifies the relative prominence of the principal spectrum, thereby intensifying the popularity bias.

DOI: 10.1145/3701551.3703579

Authors: Siyi Lin, Chongming Gao, Jiawei Chen, Sheng Zhou, Binbin Hu, Yan Feng, Chun Chen, Can Wang

Institutions: Zhejiang University, University of Science and Technology of China, Ant Group (China)

Language: English

Topic: Recommender Systems and Techniques

Subfield: Information Systems

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funders: National Natural Science Foundation of China, Universitas Brawijaya

Awards: G2727065767, G4904086664, G8033211373

OpenAlex ID: W4407953214

FWCI: 27.03

Cites: 55

Cited by: 24

Related to: 10