Don't Blame the Annotator: Bias Already Starts in the Annotation Instructions

arXiv:2205.00415v3 Announce Type: replace-cross
Abstract: In recent years, progress in NLU has been driven by benchmarks. These benchmarks are typically collected by crowdsourcing, where annotators write examples based on annotation instructions crafted by dataset creators. In this work, we hypothesize that annotators pick up on patterns in the crowdsourcing instructions, which bias them to write many similar examples that are then over-represented in the collected data. We study this form of bias, termed instruction bias, in 14 recent NLU benchmarks, …

abstract annotation arxiv benchmarks bias creators crowdsourcing cs.ai cs.cl cs.cv cs.lg dataset examples nlu patterns progress type work