Search papers, labs, and topics across Lattice.
University of Cambridge
3
0
5
SAGE achieves a statistically robust increase in user retention for a mental-health chatbot by effectively navigating the complexities of prompt optimization through agent-guided exploration.
Predicting depression severity from AI dialogue can achieve clinical-level accuracy without requiring users to fill out traditional self-report questionnaires.
A round-robin judge assignment strategy, CyclicJudge, precisely eliminates judge bias in LLM evaluation without increasing evaluation costs, offering a practical solution to a pervasive problem.
