Cohere For AI(@forai_ml) 's Twitter Profile Photo

Looking forward to attending ACL 2023 in Toronto in a few weeks.

We'll be joining our colleagues at the cohere booth on Monday, July 10. Drop by to say hello, chat about ML research, and pick up some C4AI swag. ✨

account_circle
Li Haau-Sing(@LHaausing) 's Twitter Profile Photo

1/ Paper alert 🚨 Our paper “Python Code Generation by Asking Clarification Questions” has been accepted to 🎉🥽👀🫵!

This work is a collaboration with Mo Mes, Andre Martins and Iryna Gurevych. Codegen country, let’s ride🚗~

1/ Paper alert 🚨 Our paper “Python Code Generation by Asking Clarification Questions” has been accepted to #ACL2023 🎉🥽👀🫵!

This work is a collaboration with @MohsenMes, @andre_t_martins and @IGurevych. Codegen country, let’s ride🚗~
account_circle
BGPU OFFICIAL(@bgpuofficial) 's Twitter Profile Photo

BG OFFICIAL : ทัพ ''เดอะ แรบบิท'' ปิดดีลคว้า ''เจ'' ชนาธิป คัมแบ็คไทยลีก⁣

BG Pathum United complete signing of Chanathip Songkrasin from Kawasaki Frontale

อ่านข่าว Read more bgputd.com/official/?r=ne…



BG OFFICIAL : ทัพ ''เดอะ แรบบิท'' ปิดดีลคว้า ''เจ'' ชนาธิป คัมแบ็คไทยลีก⁣ 

BG Pathum United complete signing of Chanathip Songkrasin from Kawasaki Frontale

อ่านข่าว Read more bgputd.com/official/?r=ne… 

#BGPU #บีจีปทุมยูไนเต็ด
#BGPathumUnited
#บอลไทย #ฟุตบอลไทย
account_circle
Haw-Shiuan Chang(@Haw_Shiuan) 's Twitter Profile Photo

Do you still use a CLS token in BERT-like models? Do you know we can ensemble the BERTs almost without extra cost?

🚨Single CLS token is not enough! Our new paper proposes to use multiple CLS tokens and ensemble them. 🧵[1/N]

📜arxiv.org/abs/2210.05043

Do you still use a CLS token in BERT-like models? Do you know we can ensemble the BERTs almost without extra cost?

🚨Single CLS token is not enough! Our new #ACL2023 paper proposes to use multiple CLS tokens and ensemble them. 🧵[1/N]

📜arxiv.org/abs/2210.05043
account_circle
Yijia Shao(@EchoShao8899) 's Twitter Profile Photo

What's special when using PLMs for continual learning? Our paper shows PLM's class-incremental learning performance is related to its representation ability in the learning process and the generation objective helps a lot.

🔗: arxiv.org/abs/2306.12619!

What's special when using PLMs for continual learning? Our #ACL2023 paper shows PLM's class-incremental learning performance is related to its representation ability in the learning process and the generation objective helps a lot. #NLProc 

🔗: arxiv.org/abs/2306.12619!
account_circle
Anku Rani(@anku__rani) 's Twitter Profile Photo

Submitted 2 papers to EMNLP and just realised it’s been close to 9 months of leaving well paid applied industrial job to pursue what I enjoy doing. I am attending ACL and open to PhD positions for Fall 2024. If you are recruiting and reading this, let’s meet!:)

account_circle