Search papers, labs, and topics across Lattice.
Affiliation:
2
0
5
Multilingual medical benchmarks penalize cross-lingual answer variation as model error, but real-world clinicians are sharply split on whether AI should enforce universal consistency or adapt to local cultural contexts.
By selectively amplifying updates to highly-utilized experts, Excitation rescues deep Mixture-of-Experts models from "structural confusion," enabling stable training where standard optimizers fail.
