Education
Universität des Saarlandes
2023 - current
M.Sc. in Computational Linguistics
GPA - 1.3 (4.0 on US scale)
Manipal Institute of Technology
2016 - 2020
B.Tech Computer Science and Engineering
CGPA - 9.09 / 10Relevant Work Experience
Research Assistant, Language, Computation and Cognition Lab
Dec 2023 - current
- Theoretical analysis of the strengths and limitations of all neural network building blocks
- Investigating the reasons behind length generalisation in transformers
ML Engineer, Glib.ai
Dec 2020 - Aug 2023
- Lead developer on one of the flagship products Finray, a financial statement analyser
- Implemented novel models for automated extraction of relevant tables from long documents
Intern, Samsung R&D Bangalore
Jan 2020 - Jun 2020
- Designed deep learning models for text extraction and script identification
- Curated datasets, automated parts of data cleaning for Samsung’s Alt Z Features
Intern, Novartis India Ltd
Nov 2018 - Jul 2019
- Implemented an end to end solution for comparing 2 sets of medical documents
- Designed a model to detect medical terms in a text corpus
Publications
- Huang, X., Yang, A., Bhattamishra, S., Sarrof, Y., Krebs, A., Zhou, H., Nakkiran, P., & Hahn, M. (2024). A Formal Framework for Understanding Length Generalization in Transformers. under Review at ICLR 2025
- Sarrof, Y., Veitsman, Y., & Hahn, M. (2024). The Expressive Capacity of State Space Models: A Formal Language Perspective. NeurIPS, 2024
Talks
- Formal Languages and Neural Networks seminar (FLaNN), 2024 Expressive Capacity of SSMs
- TaCoS, Student Conference, 2024 Transformers or RNNs or SSMs: Who’s more sensitive?
Key Achievements
- Smart India Hackathon 2019, Winners: Led a 6-member team to victory in a 36-hour national hackathon organized by the Government of India with 300,000 participants. Developed an AI-driven game suite based on Gardner’s Multiple Intelligences Theory to assess and enhance children’s cognitive skills.
- GE HackElt Pan India 2019, Finalists: Part of a top 10 finalist team in a national hackathon by General Electric India. Developed a predictive Hospital Appointment Management System to optimize patient wait times and provide analytics for improved hospital efficiency.
Technical Skills
- Languages: Python, C, C++
- Web Frameworks: Django, Flask
- Deep Learning Frameworks: PyTorch, TensorFlow
Last updated: 25th June 2024