Hi! I am Minbyul Jeong, a Ph.D. at Korea University under the supervision of Professor Jaewoo Kang.
I'm always passionate about solving real-world problems.
My primary goal is to enable Aritifical Intelligence to help people around the world lead better lives.
Email / CV / Google Scholar / LinkedIn / X / Github / Youtube
We discover Temporal Heads, specific attention heads primarily responsible for processing temporal knowledge through circuit analysis.
We present SysGen, a pipeline for generating system messages with better aligned assistant responses from the supervised fine-tuning dataset without system messages.
We present CHROKNOWLEDGE (Chronological Categorization of Knowledge), a novel sampling-based framework for evaluating LLMs’ non-parametric chronological knowledge.
We present MedLFQA, a benchmark dataset reconstructed using long-form question-answering. We also introduce OLAPH, a framework leverages automatic evaluation to generate synthetic preference sets that can help align the model with preferred responses.
We present Self-BioRAG, a domain-specific LLM version of Self-RAG framework.
We present ConNER, a biomedical NER training framework to enhance label consistency in document-level context.
We present BERN2, a biomedical NER and NEN framework to automatically extract biomedical entities in biomedical literature. It only spent 0.3sec per document.
We present FastGTN, a network improve scalability of graph transformations from previous version of Graph Transformer Networks (GTN).
We present GTN, a network for graph transformations to enhance node representations.
Template based on Jon Barron's website.