The Perspectivist Paradigm Shift: Assumptions and Challenges of Capturing Human Labels

arXiv:2405.05860v1 Announce Type: new
Abstract: Longstanding data labeling practices in machine learning involve collecting and aggregating labels from multiple annotators. But what should we do when annotators disagree? Though annotator disagreement has long been seen as a problem to minimize, new perspectivist approaches challenge this assumption by treating disagreement as a valuable source of information. In this position paper, we examine practices and assumptions surrounding the causes of disagreement--some challenged by perspectivist approaches, and some that remain to be addressed--as …

abstract arxiv assumptions challenge challenges cs.cl cs.cy cs.lg data data labeling human labeling labels machine machine learning multiple paradigm practices shift type