KA
Kaustubh Atey
PhD Student
Centre for Machine Intelligence and Data Science (C-MInDS),
IIT Bombay
About
I am a PhD student at the Centre for Machine Intelligence and Data Science (C-MInDS), IIT Bombay. My research studies how machine learning models can remain reliable when the data they see at test time differs from the data they were trained on. I develop methods that generalize across changing data distributions and that can estimate their own uncertainty.
Research Interests
Robust Machine Learning
Domain Generalization & Adaptation
Uncertainty Quantification & Estimation
Publications
01
DTS: Discriminant Truthfulness Scoring for Training-Free Hallucination Detection
Grounding Language Models: Learning Faithfully and Efficiently @ EMNLP, 2026
02
Mitosis Detection in the Wild: Multi-Tumor and Context-Aware Generalization in the MIDOG 2025 Challenge
arXiv preprint arXiv:2606.07368, 2026
03
Mix, Align, Distil: Reliable Cross-Domain Atypical Mitosis Classification
MICCAI (Springer), pp. 137–143, 2025
04
On the modern deep learning approaches for precipitation downscaling
Earth Science Informatics, 2023
Teaching
Teaching Assistant, IIT Bombay
- DS 203 Programming for Data Science
- DS 303 Introduction to Machine Learning
- EE 769 Introduction to Machine Learning
- EE 782 Advanced Topics in Machine Learning
Contact