Search papers, labs, and topics across Lattice.
National Institute of Informatics (NII), Japan
3
0
6
5
Current NLP metrics for "trustworthy" AI in mental health are dangerously misaligned with the actual needs of patients and practitioners.
Reward models can achieve state-of-the-art performance by critically collaborating with a rubric generator trained solely from binary preferences, eliminating the need for costly rubric annotations.
LLM-generated feedback isn't one-size-fits-all: personality dictates which elements resonate with learners, impacting acceptance and potentially learning outcomes.
