Share your thoughts, 1 month free Claude Pro on usSee more

Persuasive Dialogue on Wildfire Evacuation Simulation (WoZ-based)

92Overall Success Rate

DiPS

Updated 22d ago

Evaluation Results

Method	Links
DiPS 2026.07		92	11.2
Zero-shot 2026.07		88	11.9
RAG 2026.07		78	11.8
Random (All Personas) 2026.07		78	11.8
Random (Subset Personas) 2026.07		72	12.3
Random + RAG (Persona) 2026.07		71	12.9
Random + RAG (No Persona) 2026.07		69	12.5
Mismatched Policy 2026.07		51	13.6