I'm a PhD student in Computer Science at William & Mary, in the AURA Lab with Prof. Antonio Mastropaolo. I work on large language models for code: how to fine-tune them cheaply (LoRA/QLoRA, multi-task), and more recently what quantization does to the code they write, including its security.
Parameter-Efficient Multi-Task Fine-Tuning in Code-Related Tasks. arXiv · code
A Systematic Literature Review of Parameter-Efficient Fine-Tuning for Large Code Models (joint first author). paper · code
Quantize with Confidence? An Empirical Study of Quantization for Code Generation. arXiv
B-NER: A Novel Bangla Named Entity Recognition Dataset with Largest Entities and Its Baseline Evaluation. paper
- ALVI-Skills: add-ons for Claude Code that I use myself. One has several agents try to disprove a result before I report it, one shows my usage limits in the status line, and one caps how much of my usage a single task can spend.