Computer Vision Model Launch @ Meta Superintelligence, SAM team
Make sure to check out the launch!
Contributions:
Designing and implementing a modified version of the SuGaR algorithm, reducing runtime from 30 m to ~1 m 40 s.
Proposing and implementing pointcloud downsampling algorithm which reduces points by up to an 12% average.
Created parallelized script to improve training efficiency through data cleaning, detecting 6+% of corrupted data.
Implemented baseline for evaluation metric and created easy-to-follow jupyter notebook.
Beyond I-Con: Exploring a New Dimension of Distance Measures in Representation Learning
We extend the I-Con framework to discover new losses which achieve state-of-the-art results on clustering and supervised learning.
Contributions:
Designed, implemented, and evaluated loss functions based on the I-Con framework. Conducted analysis on experimental results.
Text-Invariant Similarity Learning
We create a new paradigm of image-pair similarity learning conditioned on text descriptions.
Contributions:
Designed and implemented image-pair similarity annotation pipeline.
Designed, implemented, and explored different experimental setups and loss functions.
Keypoint Abstraction using Large Models for Object-Relative Imitation Learning
ICRA 2025, CoRL LangRob Workshop Best Paper 2024
By utilizing priors from Vision-language models and image features from large classification models to create a novel keypoint abstraction for robot actions, we generalize effectively across object poses, camera viewpoints, and object instances with only 10 demonstrations.
Contributions:
Trained multimodal models utilizing pointcloud encoders, object-wise transformers, and vision foundation models, improving evaluation performance by 27.5%
Created and designed keypoint proposal pipeline with specialized VLM prompting SAM, furthest point sampling, mark-based prompting, RANSAC, and point cloud clustering
The Correlation of Regional Gas Prices with Unemployment
Citadel Women's Datathon Winning Report 2024
Are fluctuations in gas prices predictive of unemployment rates and how do regional differences in mass transit and gasoline production affect this trend?
Contributions:
Coming up with the initial research question + insight to use multiple additional datasets beyond the ones we were provided in the competition. Also worked on most of the stats/modelling work (Granger Causality, ADF, VAR modelling).
SketchAgent: Language-Driven Sequential Sketch Generation
CVPR 2025
We introduce SketchAgent, a novel framework for generating sequential sketches from language prompts. By leveraging large language models and diffusion models, we achieve high-quality sketch generation that captures the essence of the input text.
Contributions:
Worked on moving the LLM inference pipeline to a locally hosted version of Llama Vision 3.2 11B and Llama 3.2 90B. Learned about quantization, multi-gpu inference, and prompting techniques.
Systematic Optimization of App Generation Few Shot Learning for Large Language Models Trained on Code
Research Science Institute, Regeneron Talent Search Top 300, Acta Scientific 2023
We create an LLM in-context learning pipeline to systematically optimize (1) maximum token length of the prompt, (2) the mechanism of choosing few-shot examples, and (3) the ordering of few-shot examples to generate applications.
Contributions:
Developed the p-mRMR algorithm and prompt creation framework, designed and wrote an evaluation suite of test applications and performed manual error evaluation of generated apps.
Validating the Usage of Stable Diffusion Foundational Models on Generating Glaucoma Fundus Images in Low-Data Settings
Computer Vision Final Project
We improve stable diffusion's ability to generate high-quality fundus images of the eye, specifically for glaucoma, by finetuning on an extremely small dataset of 170 images.
Contributions:
Worked on designing the project/experiments, training the model, doing evals with resnet, and baselines (GANs)
