Semanlink - Tu Vu sur Twitter : "Enormous LMs like GPT-3 exhibit impressive few-shot performance, but w/ self-training a BERT base sized model can achieve much better results!

Impression

Recherche de Mot-clé

Recherche de Doc

Préférences...

Tu Vu sur Twitter : "Enormous LMs like GPT-3 exhibit impressive few-shot performance, but w/ self-training a BERT base sized model can achieve much better results!

Tags:

Au sujet de ce document

sl:bookmarkOf : https://twitter.com/tuvuumass/status/1514040277136293890
sl:creationDate : 2022-04-13
sl:creationTime : 2022-04-13T13:37:58Z
sl:relatedDoc :
- http://www.semanlink.net/doc/2022/04/2109_06270_strata_self_train
- http://www.semanlink.net/doc/2022/04/2203_10581_cluster_tune_bo

Infos sur le fichier

Bookmark of: https://twitter.com/tuvuumass/status/1514040277136293890

Linked From

[2109.06270] STraTA: Self-Training with Task Augmentation for Better Few-shot Learning

Tags:

2022-04-14 A propos

Documents with similar tags (experimental)

[2109.06270] STraTA: Self-Training with Task Augmentation for Better Few-shot Learning

Tags:

2022-04-14 A propos