arXiv - CSCL: "Cross2StrA: Unpaired Cross-lingual Image Captioni…" - Qoto Mastodon

arXiv - CSCL @arxiv_cscl@qoto.org

Cross2StrA: Unpaired Cross-lingual Image Captioning with Cross-lingual Cross-modal Structure-pivoted Alignment. (arXiv:2305.12260v2 [cs.CV] UPDATED)

http://arxiv.org/abs/2305.12260 #arXiv #NLProc

May 28, 2023, 03:07 · · arxiv-cscl · · ·

Sign in to participate in the conversation