I am a Pre-Doctoral Research Fellow at Microsoft PROSE. I earned my bachelor's degree in Computer Science from BITS Pilani, graduating with the Dean's Merit Scholarship (Top 1%). My research interests lie in developing machine learning solutions for Programming Languages (AI4Code) and Natural Language Processing (NLP).

My current research focuses on two main topics: (1) Using synthetic data to reliably train language models and (2) Enhancing the structured knowledge grounding capabilities of large language models (LLMs). Additionally, I am interested in handling long, unstructured inputs using LLMs and improving the interpretability of language models through symbolic reasoning. In my current work, I am fortunate to be advised by Dr. Jose Cambronero, Dr. Sumit Gulwani, Dr. Aditya Kanade, Dr. Vu Le, and Dr. Gust Verbruggen.

Previously, I interned at American Express, where I explored specialized transformers for capturing dependencies in tabular financial data. I also interned at TU Darmstadt under Prof. Iryna Gurevych, focusing on long-context question answering. During my undergraduate research at BITS Pilani, I was advised by Prof. Yashvardhan Sharma and Prof. Pratik Narang.

Publications

Asking language models how to represent data for fine-tuning
Usneek Singh, A. Singha, A. Awasthi, S. Gulwani, A. Kanade, V. Le, M. Singh, G. Verbruggen
Under review (ARR) |pdf

An Empirical Study of Validating Synthetic Data for Formula Generation
Usneek Singh, J. Cambronero, S. Gulwani, A. Kanade, A. Khatry, V. Le, M. Singh, G. Verbruggen
NAACL Findings |pdf

Comparative Analysis of Transformers for Modeling Tabular Data: A Casestudy using Industry Scale Dataset
Usneek Singh, P. Arora, S. Ganesan, M. Kumar, S. Kulkarni, and S. R. Joshi
CODS-COMAD |pdf

MFDN: Multiception Feature Distillation Network
S. Sameen*, Usneek Singh*,, and P. Narang
TENCON |pdf

Ancient Indian Murals Digital Restoration through Image InPainting
Usneek Singh, S. Maiti, A. Saini, and Dhiraj
IEEE SPIN |pdf

Multilingual Chatbot for Indian Languages
Usneek Singh, N. Vora, P. Lohia, Y. Sharma, A. Bhatia, and K. Tiwari
ICCCNT |pdf

Awards

DAAD-WISE Scholarship
DAAD, Germany
Selected among the top 100 students in India for a fully-funded research internship in Germany.

NTSE Scholarship
Govt. of India
Selected among the top 750 students out of 2 lakh applicants in the national science examination.

ACM Student Grant
ACM CODS-COMAD
Received a travel grant worth Rs. 30,000 from ACM to attend the CODS-COMAD conference

Teaching/Volunteer

Teaching Assistant, BITS Pilani
Undergraduate student assistant for two courses: Computer Programming and Data Structures and Algorithms.

Student Faculty Council, BITS Pilani
Student representative in the Department of Computer Science.

Project Leader, Nirmaan Organization
Guided a team of volunteers to teach computer skills to underprivileged youth.

BITS Pilani
2019 - 2023
Microsoft
2023-Present
American Express AI
2023
Technical University of Darmstadt
2022
CSIR_CEERI
2020