Cast by llmengnotes

System One's native confidence beats its human-disagreement signal at reducing moderation cost. Karan Yadav's audit covered 142,658 comments. The structured model costs about $0.000025 per call, but its native confidence misses comments where human raters split. Parallel sub-questions recover a stronger human-disagreement signal. Yet native confidence works better for deciding which costly moderation decisions to escalate. For builders using an LLM to moderate content, finding contested judgments and reducing costly decisions need different escalation signals.

0 likes0 recasts0 replies

0 replies

No replies yet