I am a Senior Applied Scientist at Microsoft where I work on large scale retrieval for Ads. Previously, I worked at Google Research India on improving multi-modal document understanding through LLMs, featured prominently in Google For India and Google I/O Connect. I have also contributed to novel research projects including LiT-Decoder, Node Level Differentially Private GNNs, and Low Rank Personalization.

I studied Computer Science from IIT Delhi, where I worked with Prof Mausam and Prof Parag Singla on speeding up inference in Probabilistic Graphical Models by exploiting symmetries.

Profile photo

Publications

Improving Retrieval in Sponsored Search by Leveraging Query Context Signals

M Akash Kumar, Gururaj K, Gagan Madan, Amit Singh

EMNLP 2024 | 2024 Conference on Empirical Methods in Natural Language Processing

pdf | arxiv

Sample-Efficient Personalization: Modeling User Parameters as Low Rank Plus Sparse Components

Soumyabrata Pal, Prateek Varshney, Gagan Madan, Prateek Jain, Abhradeep Guha Thakurta, Gaurav Aggarwal, Pradeep Shenoy, Gaurav Srivastava

AISTATS 2024 | 2024 International Conference on Artificial Intelligence and Statistics

pdf | arxiv

A Study of Autoregressive Decoders for Multi-Tasking in Computer Vision

Lucas Beyer, Bo Wan, Gagan Madan, Filip Pavetic, Andreas Steiner, Alexander Kolesnikov, André Susano Pinto, Emanuele Bugliarello, Xiao Wang, Qihang Yu, Liang-Chieh Chen, Xiaohua Zhai

pdf | arxiv

Weakly Supervised Information Extraction from Inscrutable Handwritten Document Images

Sujoy Paul, Gagan Madan, Akankshya Mishra, Narayan Hegde, Pradeep Kumar, Gaurav Aggarwal

ICDAR 2023 | 2023 International Conference on Document Analysis and Recognition

pdf | arxiv

Media coverage: TechCrunch | Google For India | Google Blog

Node-Level Differentially Private Graph Neural Networks

Ameya Daigavane, Gagan Madan, Aditya Sinha, Abhradeep Guha Thakurta, Gaurav Aggarwal, Prateek Jain

ICLR 2022 Pair2Struct Workshop | 2022 Privacy, Accountability, Interpretability, Robustness, Reasoning on Structured Data Workshop at ICLR

pdf | arxiv | code

🏆 BOSS Award 2018 Block Value Symmetries in Probabilistic Graphical Models

Gagan Madan, Ankit Anand, Mausam, Parag Singla

UAI 2018 | 2018 Conference on Uncertainty in Artificial Intelligence

pdf | arxiv | code