Nikhil Prakash
(he/him/his)
PhD Student
Research interests
- Machine learning
- Natural language processing
- Human-computer interaction
Education
- BS in Telecommunication Engineering, RV College of Engineering, Bangalore, India
Biography
Nikhil Prakash is a PhD student at the Khoury College of Computer Sciences. He earned his bachelor’s in telecommunication engineering from RV College of Engineering in Bangalore, India. He is affiliated with the Bau Lab.
Prakash’s research area is the interpretability of large language models, and utilizing the insights gained from probing experiments to develop better Human-AI Collaboration techniques for Human-Centered AI systems. He intends to pursue this research direction by amalgamating theories and concepts from Machine Learning and Human-Computer Interaction domains. His faculty advisor is Professor David Bau.
He has presented his work at COMSNETS 2019, NeurIPS 2020, and HCOMP 2021.
Outside of work, Prakash enjoys playing chess and table tennis, watching documentaries, anime, and learning about culture and history.
Recent publications
-
Fine-Tuning Enhances Existing Mechanisms: A Case Study on Entity Tracking
Citation: Nikhil Prakash, Tamar Rott Shaham, Tal Haklay, Yonatan Belinkov, David Bau. (2024). Fine-Tuning Enhances Existing Mechanisms: A Case Study on Entity Tracking ICLR. https://openreview.net/forum?id=8sKcAWOf2D -
iClarify – A Tool to Help Requesters Iteratively Improve Task Descriptions in Crowdsourcing
Citation: Nouri Z., Prakash N., Gadiraju U., Wachsmuth H. (2021b). “iClarify - A tool to help requesters iteratively improve task descriptions in crowdsourcing,” in Proceedings of the 9th AAAI Conference on Human Computation and Crowdsourcing (HCOMP). -
Conceptualization and Framework of Hybrid Intelligence Systems
Citation: Nikhil Prakash and Kory W. Mathewson, "Conceptualization and Framework of Hybrid Intelligence Systems", arXiv:2012.06161 (2020).