Honorable mention paper examines over-reliance on LLM explanations in hate speech moderation
The ACM abstract says the study evaluates how LLM-generated explanations support human moderators detecting hate speech, particularly when discriminatory intent is implicit or nuanced.
TLDR
A post says Bing Tuo traveled from Australia to Washington, DC, to present the honorable mention paper at HCOMP 2026. Its title argues that easy-to-read LLM explanations drive over-reliance in hate speech moderation. The ACM abstract says the study evaluates different types of explanations as aids for human moderators.
Combined views
167
2 Sources, first seen 1h ago
