Sanket Vaibhav Mehta, Ph.D.
@sanketvmehta
Research Scientist @GoogleDeepMind | Ph.D. @LTIatCMU @SCSatCMU @CarnegieMellon | Past @GoogleAI, @AdobeResearch, @IITRoorkee
ID:1189682563
https://sanketvmehta.github.io/ 17-02-2013 14:23:58
199 Tweets
713 Followers
1,1K Following
“How to achieve successful continual learning?” Christopher Manning keynote at #EMNLP2023
Join us at #NeurIPS2023 , where we present our #JMLR paper shedding light on the pivotal role of pre-training in successful continual learning!
Thanks Clara Na, Amanda Bertsch and Adithya Pratapa for organising the first #RUNLP session at #EMNLP2023 yesterday! #RUNLP ™️ by Pratyush Maini! It was a fun, and looking forward to more sessions at future conferences!
Zachary Lipton any plans for #NeurIPS2023 runconference?
High-quality data is a key to successful pretrain/finetuning in the GPT era, but manual data curation is expensive💸 We tackle data quality challenges involving large models and datasets with ScAlable Meta leArning (SAMA) #NeurIPS2023 💫
Arxiv: arxiv.org/abs/2310.05674
🧵 (1/n)