Currently, I am a research intern at the Vision and AI Lab (VAL) at the Indian Institute of Science, working under Prof. Venkatesh Babu on object-centric representation learning. I was also a research intern at the University of California, Berkeley, working with Prof. Prasad Raghavendra on mechanistic interpretability of neural networks.
Previously, I conducted research at the CDSAML Lab at PES University under Prof. Shylaja S S, focusing on hypergraph-based models for sentiment analysis. I am also currently supported by the Junior Research Fellowship at PES University.
Keywords: Representation Learning, Mechanistic Interpretability, Vision Language Models
I'm always open to collaboration, so feel free to reach out if you'd like to connect or work together!
From
Graphs to Hypergraphs: Enhancing Aspect-Term Sentiment
Analysis via Multi-Level Relational Modeling
, Amit Padegal*, Madhav Kashyap,
Ashwini. M. Joshi, Shylaja S. S.
ACL'26 pdf
Quantifying Modality Contributions via Disentangling Multimodal
Representations
Amit Padegal*, , Namitha Rayasam*,
Nidhi
Shekhar, Surabhi Narayan.
FLAIRS'26 (🏆 Best Paper Award) pdf
Lost in the Slots: Revisiting
Object-Centric Representations
in the era of Foundation Models
Priyam Dey, Aditya Sahdev, , R.
Venkatesh Babu
NeurIPS'26 (Under Review)
Accepted to UCSD to pursue graduate studies!
Our paper on Aspect-based Sentiment Analysis using Hypergraphs has been accepted at ACL 2026!
Presented our work on Vision-Language models at FLAIRS 2026, where it won the Best Paper Award!
Began a research internship at the Vision and AI Lab, Indian Institute of Science, under the guidance of Prof. R. Venkatesh Babu.
Our paper on Quantifying modality contribution in VLMs has been accepted at FLAIRS 2026!
Started a research internship at the University of California, Berkeley, under the guidance of Prof. Prasad Raghavendra.
Was awarded the Junior Research Fellowship at PES University.
Completed a research internship at the CDSAML Lab, PES University, under the guidance of Prof. Ashwini M Joshi and Prof. Shylaja S S
