Scholar
Sharut Gupta
Google Scholar ID: xxiTm5EAAAAJ
PhD at MIT CSAIL
Citations & Impact
Citations
1,076
H-index
14
i10-index
18
Publications
20
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- Published 'Representation Guidance for Diffusion Models' at NeurIPS 2025, with a short version appearing as an oral presentation at the ICML'25 FM4LS Workshop. Published 'Learning Disentangled Multimodal Representations' at ICLR 2025. Awarded the MathWorks Engineering Fellowship. Published 'In-Context Symmetries' and 'Understanding the Role of Equivariance in Self-supervised Learning' at NeurIPS 2024, with a short version appearing as an oral presentation at the NeurIPS'24 SSL Workshop. Published three papers at ICLR 2024: 'Context is Environment', 'Structuring Representation Geometry with Rotationally Equivariant Contrastive Learning', and 'Removing Biases from Molecular Representations via Information Maximization'.
Research Experience
- Spent time at Meta SuperIntelligence Labs (Meta AI) and Google DeepMind. Co-organized several seminars and academic events, such as the ML Tea Seminar.
Education
- Earned a dual Bachelor's and Master's degree in Mathematics and Computer Science from the Indian Institute of Technology, Delhi (IIT Delhi), where he was mentored by Prof. Yoshua Bengio for his thesis. Currently pursuing a Ph.D. at MIT CSAIL, advised by Prof. Phillip Isola and Prof. Stefanie Jegelka.
Background
- A fourth-year Ph.D. student at MIT CSAIL, with research interests broadly in Machine Learning, specifically around multi-modal representation learning, robustness, and out-of-distribution generalization. Recently, he has been thinking about the role of data from diverse modalities and its implications for unified architectures.
Miscellany
- Attended the Aspen Meeting on Foundation Models. Featured in the MIT CSAIL Alliances Student Spotlight and joined the MIT CSAIL Alliances podcast to discuss recent work. Organized the WiDS Cambridge Datathon 2024. Served on the executive committee for the MIT Graduate Application Assistance Program (GAAP).
Co-authors