Ji-Woo Song

I am pursuing an M.S. in Artificial Intelligence, taking a comprehensive AI curriculum across core areas and developing a strong focus on NLP. Current GPA: 4.5/4.5.

Conducted machine learning research and AI-related projects, graduating with a major GPA of 4.08/4.5 and an overall GPA of 4.00/4.5. In addition to AI, gained broad experience in computer science, including web development, SQL, Linux, and data analysis.

Conducting research on NLP, LLM, and hallucination mitigation in multi-modal (VLM) models.

As a graduate researcher at the NLP Lab, I focus on enhancing the performance of Large Language Models (LLMs) and multi-modal Vision-Language Models (VLMs) by mitigating hallucination issues. My work involves fine-tuning pre-trained models, optimizing retrieval-augmented generation (RAG) methods, and developing novel approaches to improve model reliability.

Conducted research on feature selection using genetic algorithms to improve machine learning-based sarcopenia prediction.

Focused on applying genetic algorithms for feature selection in sarcopenia prediction models. Processed survey data from the KLoSA dataset, reducing features while improving model performance. Evaluated logistic regression, XGBoost, and random forest classifiers to assess the effectiveness of selected features.