arXiv - CSCL: "IMAGINATOR: Pre-Trained Image+Text Joint Embeddin…" - Qoto Mastodon

arXiv - CSCL @arxiv_cscl@qoto.org

IMAGINATOR: Pre-Trained Image+Text Joint Embeddings using Word-Level Grounding of Images. (arXiv:2305.10438v1 [cs.CL])

http://arxiv.org/abs/2305.10438 #arXiv #NLProc

May 21, 2023, 03:07 · · arxiv-cscl · · ·

Sign in to participate in the conversation