Abhinav Ramesh Kashyap
ASUS-AICS National University of Singapore WING NUS

I am a Research Scientist at ASUS-AICS. I am also a part-time fifth-year Ph.D. student advised by Prof Min-Yen Kan. I closely collaborate with Devamanyu Hazarika and Soujanya Poria. My research focuses on Natural Language Processing.
Specifically I make NLP models robust under different domains, also called Domain Adaptation. I am interested in using unsupervised data from different domains for domain adaptation. My recent interest focuses on using parameter efficient learning methods for domain adaptation of Large Language Models.
In AICS, I work on Clinical Natural Language Processing. If you want to know more about this reserach and for collaborations, email me.
Previously,
I have done an internship at Amazon, with the Alexa AI team. I completed my Master’s degree from the National University of Singapore. His master’s thesis involved extractive summarization specific to restaurant reviews. I was part of the Sensor Enhanced Social Media(SeSaMe) lab with Prof. Mohan Kankanhalli
News
Jan 10, 2023 |
Our paper “UDAPTER: Efficient Domain Adaptation using Adapters” has been accepted for EACL’23. This paper aims to make domain adaptation more efficient. This is a joint work with Bhavitvtya Malik. Code, paper and everything nice, coming soon. |
---|---|
Jan 10, 2023 |
Our Survey paper “Scientific Document Processing: Challenges for Modern Learning Methods” is accepted at IJDL. |
Oct 17, 2022 |
Excited to join ASUS-AICS. I will be working on Clinical NLP. |
Aug 31, 2022 |
Passed my Thesis Proposal at NUS |
May 25, 2022 |
Presented our Paper “So Different Yet So Alike! Constrained Unsupervised Text Style Transfer” at ACL’22 as an oral presentation |
May 13, 2022 |
Presented our paper “So Different Yet So Alike! Constrained Unsupervised Text Style Transfer” at SSNLP’22 |
Apr 22, 2022 |
Recognized as an Outstanding Reviewer for ICLR 2022 |
Apr 1, 2022 |
Wrapped up my internship at Amazon. |
Feb 24, 2022 |
Our work “So Different Yet So Alike! Constrained Unsupervised Text Style Transfer” is accepted at ACL. |
Oct 1, 2021 |
Started as an Applied Scientist Intern @ Amazon Science. Excited to work with Mahdi Namazifar |