Annotation alignment: Comparing LLM and human annotations of conversational safety.

Rajiv Movva, Pang Wei Koh, Emma Pierson

Browse the full EMNLP paper archive.