arXiv - CSCL: "Merging Experts into One: Improving Computational…" - Qoto Mastodon

arXiv - CSCL @arxiv_cscl@qoto.org

Merging Experts into One: Improving Computational Efficiency of Mixture of Experts. (arXiv:2310.09832v3 [cs.CL] UPDATED)

http://arxiv.org/abs/2310.09832 #arXiv #NLProc

Nov 23, 2023, 03:18 · · arxiv-cscl · · ·

Sign in to participate in the conversation