Projects

Video QA and RAG system for structured curation and grounded retrieval over long multimodal content.

Structure-aware multimodal chunking pipeline for grounded retrieval and long-document QA.

Science-domain Korean LLM and RAG pipeline for document-grounded industrial QA.

Insurance document QA pilot combining LLM-based retrieval with production-oriented application development.

Multilingual medical consultation system with LLM chatbot and production-facing admin workflows.

Consortium-scale multilingual foundation model contribution spanning data pipelines and evaluation.

Korean-adapted Gemma development with Korean-focused data and evaluation.

Academic Korean LLM development through post-training and benchmark evaluation.

GPT-based code-generation assistant for smart-factory PLC workflows in collaboration with Hyundai Mobis.

Enterprise Korean LLM adaptation and evaluation for generative AI applications.

Adapter-based architecture for Table QA and neural-symbolic Korean sentence analysis; reduced QA model count by 50% and grounded later work on structured evidence.

Sentence analysis and prosodic boundary prediction for Korean TTS quality improvement (92.13%).

Korean phoneme sequence generation for higher-quality TTS systems, deployed in KT GiGA Genie.

Korean phoneme-based navigation TTS system developed with Navice, deployed in BMW and Mercedes-Benz vehicles.

AI-based online judge system deployed in the PNU SW curriculum with NextStage.

Geoscience AI system for mineral location prediction and domain-specific feature engineering.