AIBullisharXiv – CS AI · 10h ago7/10
🧠
Evidence Over Plans: Online Trajectory Verification for Skill Distillation
Researchers introduce SPARK, a framework that verifies AI agent skills through direct environment interaction rather than relying on pre-written plans. The Posterior Distillation Index (PDI) metric ensures skills are grounded in actual task evidence, producing student models that match or exceed human-written skills while reducing inference costs by up to 1,000x.