PersonaGym
Benchmark for persona agent capabilities
Episodes
-
#2186: The AI Persona Fidelity ChallengeAdvanced LLMs dominate benchmarks but fail at staying in character—especially when asked to play morally complex or antagonistic roles. What does t... -
#2133: Engineering Geopolitical Personas: Beyond CaricaturesHow to build LLMs that simulate state actors with strategic fidelity, not just surface mimicry.