Undergraduate Research Assistant
Conducted Fine-tuning research on Large Language Models (LLMs) like CodeLlama and MagicCoder, using 8 different code representations, 3 standard Automated Program Repair (APR) datasets, LoRA, CoT and vLLM techniques to analyze the impact of various factors on the model’s performance in APR tasks. Achieved superior performance in APR tasks, outperforming.