Hanyong Lee
AI researcher

2026 / Seoul
CAU AutoML Lab

I build AI systems that stay useful when things get messy.

Trustworthy NLP, research systems,
and evidence-aware evaluation.

PhD researcher in Artificial Intelligence
at Chung-Ang University.

Publications GitHub Scholar

Language. Evidence. Execution.Explore the work

01 / Selected research

All publications

2025Findings of NAACL 2025

BitAbuseA Dataset of Visually Perturbed Texts for Defending Phishing Attacks

H. Lee, C. Lee, Y. Lee, J. Lee

BitAbuse Figure 1: character-word association graph and vowel subgraphs from the BitCore dataset — Lee et al., 2025 · Fig. 1 / BitCore associations

2025Computers, Materials & Continua

OCR-Assisted Masked BERTfor Homoglyph Restoration towards Multiple Phishing Text Downstream Tasks

H. Lee, Y.-C. Park, J. Lee

OCR-Assisted Masked BERT Figure 2: data preparation, pre-training, and fine-tuning for text restoration — Lee et al., 2025 · Fig. 2 / Restoration framework · CC BY 4.0

2026Neurocomputing

Joint compression & fine-tuningA survey of large language models: methods, toolchains, and open challenges under resource constraints

J. Kim, H. Lee, J. Lee

02 / System

Research infrastructure

AutoLabOS

Research should
not be a black box.

An operating system for
autonomous research.

AutoLabOS turns research execution into a governed, inspectable, and resumable workflow.

View system GitHub

The research loop

01Literature
02Research map
03Hypothesis
04Experiment plan
05Execution
06Review gate
07Evidence
08Manuscript

Inspectable by design.Checkpoint / Review / Resume

03 / Notes & updates

2026Survey on joint compression and fine-tuning published in Neurocomputing 2025BitAbuse at Findings of NAACL 2025OCR-Assisted Masked BERT published in Computers, Materials & Continua

About

I am a PhD researcher in Artificial Intelligence at Chung-Ang University. I work where language, reliability, and research systems meet.