Search papers, labs, and topics across Lattice.
Harvard T.H. Chan School of Public Health, Harvard Medical School
3
0
5
LLMs miss critical early-stage patient interactions, providing self-care advice in only 75% of baseline scenarios compared to none under structured instructions.
Tying every view of a digital artifact to its version can boost knowledge-work agent performance by over 12 points on critical tasks.
Despite improvements from prompting, patient-facing LLMs still misjudge 66.5% of the time when more information is needed before recommending care.
