AI RESEARCH
From Prompt Risk to Response Risk: Paired Analysis of Safety Behavior of Large Language Model
arXiv CS.CL
•
ArXi:2604.26052v1 Announce Type: new Safety evaluations of large language models (LLMs) typically report binary outcomes such as attack success rate, refusal rate, or harmful/not-harmful response classification. While useful, these can hide how risk changes between a user's input and the model's response. We present a paired, transition-based analysis over 1250 prompt-response records with human-provided labels over four harm categories (Hate, Sexual, Violence, Self-harm) and ordinal severity levels aligned with the Azure AI Content Safety taxonomy.