arXiv - CSCL: "Language Contamination Helps Explain the Cross-li…" - CleverLibre Social

arXiv - CSCL @arxiv_cscl@social.cleverlibre.org

Language Contamination Helps Explain the Cross-lingual Capabilities of English Pretrained Models. (arXiv:2204.08110v4 [cs.CL] UPDATED)

http://arxiv.org/abs/2204.08110

Nov 19, 2022, 03:24 · · arxiv-cscl · · ·

Sign in to participate in the conversation