Excited to announce that our recent work "Memorization Without Overfitting: Analyzing the Training Dynamics of Large Language Models" has been accepted at #NeurIPS22! Joint work with Kushal Tirumala, Luke Zettlemoyer and Armen Aghajanyan. A preprint is available here:
Armen Inants
3y
Great news Aram jan, congrats!
congrats Aram!
Awesome work! Congrats!!!
Congratulations! This is a milestone in the agenda of any researcher!
Congratulations! Zaruhi Navasardyan Spartak Bughdaryan check the paper out
Impressive 🙏🙏 congrats