Algorithmic Arbitrariness in Content Moderation

arXiv:2402.16979v1 Announce Type: cross
Abstract: Machine learning (ML) is widely used to moderate online content. Despite its scalability relative to human moderation, the use of ML introduces unique challenges to content moderation. One such challenge is predictive multiplicity: multiple competing models for content classification may perform equally well on average, yet assign conflicting predictions to the same content. This multiplicity can result from seemingly innocuous choices during model development, such as random seed selection for parameter initialization. We experimentally demonstrate …

abstract arxiv challenge challenges classification content moderation cs.cy cs.lg cs.si human machine machine learning moderation multiple online content predictions predictive scalability type