PhD Student, EECS · MIT CSAIL
I'm a PhD student in EECS at MIT CSAIL, co-advised by Antonio Torralba and William T. Freeman. My research focuses on advancing self-supervised learning and vision-language models.
I completed my B.S. in Mathematics and Computer Science at MIT, where I worked with amazing mentors Deva Ramanan and Shu Kong at CMU's Argo AI Center, as well as Abhinav Gupta and Victoria Dean at CMU's Robotics Institute.
Beyond research, I'm active in the math olympiad community as a former contestant (IMO Bronze 2017, EGMO and BMO Gold 2016). I also train students, design problems, and I served as a deputy leader and observer at IMO and EGMO.
A large-scale multimodal, multilingual dataset of 30K+ olympiad-level problems from competitions worldwide, paired with a benchmark that measures both step-by-step reasoning in generative models and mathematical retrieval in embedding models.
A unified framework that generalizes loss functions in representation learning, exposing connections across methods and achieving state-of-the-art results in unsupervised image classification on ImageNet-1K.
A benchmark evaluating negation understanding in vision-language models reveals performance limitations, with targeted improvements increasing recall by 10% and accuracy by 40%.
This study explores weight balancing techniques like L2-normalization, weight decay, and MaxNorm to address bias in long-tailed recognition, achieving state-of-the-art results across five benchmarks by balancing classifier weights for rare and common classes.
This work combines visual data with contact audio to enhance manipulation in contact-rich tasks, leveraging high-frequency tactile signals from microphones to outperform single-modality approaches.
This work introduces a continual learning approach for long-tailed recognition, using a Mean-Shift module and Supervised Contrastive loss to improve feature learning and expedite finetuning across time periods, achieving state-of-the-art performance.
MIT Department of Mathematics · Sep 2022 – Jan 2025
I teach two weekly recitation sessions to help clarify challenging topics for 38 students and develop weekly handouts and problem sets for a larger group of 180 students. I also coordinate a team of five TAs and three Graders. I was honored to be nominated by my students for the Teaching Awards.
KAUST · Jun 2025
MIT EECS Department · Jan 2024 – May 2024
Supported professors in organizing technical materials on ML topics, conducted weekly recitations, lab sessions, and hosted office hours for student learning support.
Deputy Leader and Observer @ IMO & EGMO · 2019–2023
Trained students in combinatorics, number theory, algebra, and geometry for the International Math Olympiad (IMO), focusing on advanced problem-solving skills. Additionally, contributed by suggesting problems for exams for team selection tests.