This research introduces a novel approach for LLMs to identify and address knowledge gaps during multi-turn conversations, creating more natural and personalized interactions.

• Identifies the persona knowledge gap as a key challenge in maintaining coherence during conversations
• Presents the Conversation Preference Elicitation and Resolution (CPER) framework to detect and resolve these gaps
• Demonstrates effectiveness in mental health support conversations, showing particular promise for sensitive support contexts
• Provides a path toward more contextually aware AI assistants that can gracefully handle uncertainty

For support applications, this research offers a structured way to maintain conversational coherence while gathering necessary user information, significantly improving the quality of AI-driven support experiences.

From Guessing to Asking: An Approach to Resolving the Persona Knowledge Gap in LLMs during Multi-Turn Conversations