TY  - CONF
AU  - Schuhmann, Christoph
AU  - Vencu, Richard
AU  - Beaumont, Romain
AU  - Kaczmarczyk, Robert
AU  - Mullis, Clayton
AU  - Katta, Aarush
AU  - Coombes, Theo
AU  - Jitsev, Jenia
AU  - Komatsuzaki, Aran
TI  - LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs
M1  - FZJ-2022-00923
SP  - 5 p.
PY  - 2021
AB  - Multi-modal language-vision models trained on hundreds of millions of image-textpairs (e.g. CLIP, DALL-E) gained a recent surge, showing remarkable capability toperform zero- or few-shot learning and transfer even in absence of per-sample labelson target image data. Despite this trend, to date there has been no publicly availabledatasets of sufficient scale for training such models from scratch. To address thisissue, in a community effort we build and release for public LAION-400M, adataset with CLIP-filtered 400 million image-text pairs, their CLIP embeddingsand kNN indices that allow efficient similarity search
T2  - NeurIPS Workshop Datacentric AI
CY  - 14 Dec 2021 - 14 Dec 2021, online (online)
Y2  - 14 Dec 2021 - 14 Dec 2021
M2  - online, online
LB  - PUB:(DE-HGF)8
UR  - https://juser.fz-juelich.de/record/905696
ER  -