Atharva Kulkarni
Hello ! I am a second year Masters of Language Technologies (MLT) student in the Language Technologies Institute, School of Computer Science at Carnegie Mellon University, advised by Barnabás Póczos. My research interests lie at the intersection of machine learning and natural language processing, with a particular focus on:
- Analyzing and enhancing generalization, robustness, and fairness of present-day neural networks (particularly LLMs).
- Learning with limited data, active learning, data valuation algorithms, and data-centric machine learning.
- Parameter efficient machine learning.
Currently, I am working on information theoretic measures for improving fairness and robustness of ML/NLP systems. I’ve also been fortunate to collaborate with several esteemed facutly at CMU, such as Graham Neubig, Aditi Raghunathan, Ameet Talwalkar, Emma Strubell, and David R. Mortensen, on various research projects that analyze and improve generalization and efficiency of neural networks. For the summer of 2023, I interned with Apple Research (Siri), investigating the emergent capabilities of LLMs for dialog applications.
Before coming to CMU, I was a Predoctoral Researcher / Research Associate with Prof. Tanmoy Chakraborty at the Laboratory for Computational Social Systems (LCS2), IIIT Delhi working on various research projects in NLP, multimodal machine learning, social computing, and conversational systems. My research has been published at top NLP/ML conferences such as ACL, EMNLP, EACL, SIGKDD, and IJCAI.
You can learn more about my publications here. You can find my detailed CV here.
I’m eager to connect with my academic peers! If our research interests align (or diverge) in intriguing ways, I’d be delighted to explore potential collaborations or simply exchange ideas! Additionally, I’m also looking for research internship opportunities for Summer 2024 to work on data-centric machine learning or generalization / fairness / efficiency of LLMs. Please feel free to reach out via email, if there is a good fit!
News
Feb 2024 | Long standing work on Multitask Learning for Worst-Group Generalization got accepted to TMLR 2024! |
---|---|
Jan 2024 | Summer 2023 internship work at Apple ML Research (Siri) on synthetic data generation for few-shot DST got accepted to EACL 2024 main conference. See you in Malta! |
Oct 2023 | Work on Wuggpt and Facutal Error Correction accepted to EMNLP main conference and Findings, respectively. |