Author
Ziv Ben‐Zion
Recent research
- AI & ComputingOpen access
Inference-Time Self-Guidance for Conversational Safety in Delusion-Related Interactions
Background: Large language models (LLMs) are increasingly used in psychologically sensitive conversations, where repeated affirmation may reinforce delusion-related beliefs and associated risk. Inference-time self-guidance, in which a model generates conversation-specific guidanc...
- AI & ComputingOpen access
Inference-Time Self-Guidance for Conversational Safety in Delusion-Related Interactions
Background: Large language models (LLMs) are increasingly used in psychologically sensitive conversations, where repeated affirmation may reinforce delusion-related beliefs and associated risk. Inference-time self-guidance, in which a model generates conversation-specific guidanc...