The goal of this project was to provide an open source implementation of the Moment Matching framework by Salimans et al. to distill image diffusion models, releasing training code and weights that were previously unavailable. Unlike consistency distillation, this is fundamentally a distribution matching problem, allowing the student to match a higher performing stochastic teacher. The core objective is to ensure the conditional expectation of clean data given a noisy observation remains identical between the true data and the distilled model. We use an alternating optimization setup where an auxiliary denoising model estimates this exact expectation, denoted as $\mathbb{E}_{g}[\tilde{\mathbf{x}} \mid \mathbf{z}_s]$, where $g$ is the generator distribution, $\tilde{\mathbf{x}}$ is the generated data, and $\mathbf{z}_s$ is the noisy observation at step $s$. By matching these moments, our 8 step distilled model reaches an FID of 1.4. Training code, checkpoints, and the full implementation are on GitHub.

Building on the work of Heek et al., this project provides an open source implementation and public weights for Multistep Consistency Distillation to accelerate diffusion models on ImageNet. We translated their original Variance Preserving framework into the Variance Exploding parameterization, specifically the EDM framework by Karras et al. Instead of uniform edge sampling, we sampled the noise parameters for the teacher and student using a log normal distribution, which significantly improved convergence. In a neat result, our 8 step student model successfully matches the deterministic teacher at an FID of 2.3. The code and model weights are available here.

This project addresses the deployment and computational hurdles of modern biomolecular prediction and design models like AlphaFold 3, Boltz 2, Chai 1, and RFdiffusion. To make these tools more accessible, we developed optimized open source Singularity containers for both x86 and ARM architectures. Because multiple sequence alignment generation is highly resource intensive, we also built OmniFold. This platform runs these models simultaneously, recycles alignments across algorithms to save compute, and automatically generates a unified quality control report. All containers and code are publicly available to help researchers bypass infrastructure bottlenecks. A preprint is available here, and the OmniFold code is on GitHub.

Developed an optimized open-source implementation of AlphaFold 2 & 3 in joint work with Dr. William Lai at the Epigenomics Facility at Cornell University that addresses a fundamental challenge in computational structural biology: resource inefficiency. The AlphaFold workflow has roughly 75% of runtime consumed by CPU-intensive multiple sequence alignment (MSA) generation, with GPU resources only required for the final structure prediction phase. Our solution separates these phases within a single Open OnDemand (OOD) instance. This work was presented at the Global Open OnDemand Conference at Harvard University in March 2025. Our work has been adopted by several major HPCs in the United States and around the world. The code is available here.

This project is part of my M.Sc thesis under the supervision of Dr. Soundar Kumara and Dr. Peter Butler at Penn State. It aims to extract insights from biomedical literature by generating word embeddings to create weighted word networks. Due to the advances in large language models, biomedical named entity recognition has become much more reliable. Hence, depending on the research question, we use keywords to extract relevant information from these embedding models while carefully selecting the data sources for training. When limiting our models to literature published before a specific year, we discovered mentions of entities in our network that were not officially recognized until later, providing new avenues for the research questions we are investigating.

During my summer research internship at IIT Guwahati under Professor Swarup Bag, I developed a university timetable generator using a parent-centric recombination operator in the Genetic Algorithm. The goal was to synchronize room constraints, subject schedules, and faculty availability/preferences, ensuring that labs or lectures were not scheduled consecutively. The algorithm began with a random timetable set, assessed its fitness, executed crossovers, and iteratively improved the solutions. The final product was designed for maximum efficiency in timetable structuring. I have shared the code on my GitHub for both academic and real-world applications.