JudgeMyAISOPO: Socratic Guided Policy Optimization for Span-Level Hallucination Detection
SOPO: Socratic Guided Policy Optimization for Span-Level Hallucination Detection
Tag: SOPO
SOPO: Socratic Guided Policy Optimization for Span-Level Hallucination Detection
A 60-second tour of VideoEval scoring 10 AI-generated clips: prompt adherence, color saturation, motion ghosting, pass/fail verdicts, and the human review queue. Real demo, real numbers.