SG-Bench: Evaluating LLM Safety Generalization Across Diverse Tasks and Prompt Types

Devs

SG-Bench: Evaluating LLM Safety Generalization Across Diverse Tasks and Prompt Types | Read Paper on Bytez