arXiv - CSCL: "Stack Attention: Improving the Ability of Transfo…" - Qoto Mastodon

arXiv - CSCL @arxiv_cscl@qoto.org

Stack Attention: Improving the Ability of Transformers to Model Hierarchical Patterns. (arXiv:2310.01749v2 [cs.CL] UPDATED)

http://arxiv.org/abs/2310.01749 #arXiv #NLProc

Jan 25, 2024, 03:19 · · arxiv-cscl · · ·

Sign in to participate in the conversation