Research

Research Areas연구 분야

Four lines of work, one thread: systems that make data fast, recoverable, and trustworthy.

2024 – present · funded

nlp-metadata · 권리 메타데이터

NLP & Rights Metadata 공유저작물 권리 메타데이터 추출

67-field metadata schema · 144k-work KOGL corpus

A deployed NER + LLM “division of labor” pipeline for Korean public-domain rights documents: a multi-provider OCR fallback chain feeds a fine-tuned KLUE-RoBERTa-Large NER model (26 B-I-O labels) and schema-guided LLM extraction running concurrently, and an LLM consolidation arbiter produces a unified 67-field schema with per-field provenance (AGREED / CONFLICT / LLM_ONLY / NER_ONLY / MISSING) and OCR-grounded Korean evidence.

Year 2 adds image + text multimodal analysis — generative VLMs and SigLIP 2 + FAISS similarity — over a 144k-work KOGL corpus. The work is funded by the MCST/KOCCA copyright R&D program, in a consortium with Muhayu and HM Company serving the Korea Copyright Commission (Gongu Madang) and the Korea Culture Information Service Agency (KOGL).

Architecture diagram: OCR feeds concurrent LLM extraction and Korean NER, an LLM arbiter consolidates both into a unified 67-field metadata schema — NER–LLM consolidation architecture for rights-metadata extraction (deployed pipeline)

Voice Phishing Detection 한국어 보이스피싱 탐지

KorCCVi — the first labeled Korean voice-phishing dataset

South Korea’s National Police Agency counts 260,694 cumulative vishing incidents from 2016 to 2025, with KRW 6,015.3 billion (≈ USD 4.11 B) in damage. Against that backdrop the lab built KorCCVi, the first labeled Korean voice-phishing transcript dataset, and progressed from ML/DL text classifiers and KoBERT to an attention-based CNN-BiLSTM (F1 0.9966), federated learning, and multilingual back-translation with SMOTE augmentation.

The current line is multimodal (audio + text) and privacy-preserving: lightweight acoustic models (≤223K parameters; 99.59% accuracy with eGeMAPS features) and channel-robust Wav2Vec2.0 with EMA call-level aggregation (F1 ≈ 0.956–0.96 under GSM codec-matched evaluation) — efficient enough for real-time detection; see the papers for benchmark details. A five-year MSIT proposal extends this to deepfake benchmarks (KorCCVi-DF), adversarial defense in the LLM era, large audio-language models, and federated learning with differential privacy.

Pipeline diagram: input audio stream through preprocessing, eGeMAPS/MFCC/Wav2Vec2.0 feature extraction, a lightweight MLP classifier, and EMA call-level aggregation to a vishing alert — fully on-device — On-device, privacy-preserving detection pipeline — acoustic features to a call-level alert (IEEE Access line)

2010 – 2020

forensics · 디지털 포렌식

Digital Forensics 플래시 저장장치 · SQLite 디지털 포렌식

10+ M.S. theses on flash & SQLite recovery (2010–2020)

A decade of digital forensics on flash storage: recovery of deleted and wiped files on NAND flash, anti-forensics via block permutation to prevent recovery, wiping-evidence acquisition, selective recovery under the SSD TRIM command, YAFFS2 file-system metadata management, and circumstantial-evidence recovery for forensic timeline analysis.

The same line extends to databases in the field — SQLite deleted-record recovery, journaling-of-journal-based SQLite file recovery, and message recovery from instant messengers. Altogether this area produced more than ten master’s theses between 2010 and 2020.

Diagram of block-mapping and log-page-mapping tables being restored from map blocks and pruned by a consistency check after a crash — Mapping-table restoration and consistency check after power failure (Electronics 2021)

Flash Storage & FTL 플래시 메모리 저장 시스템 & FTL

FAST FTL — 712 citations (ACM TECS 2007)

The lab co-created the widely cited FAST flash translation layer (“A log buffer-based flash translation layer using fully-associative sector translation”, ACM TECS 2007 — cited 700+ times) and the LSTAFF/LSTAFF* system software for large-block flash. The work spans FTL design and surveys, hybrid address mapping, buffer replacement, flash-aware B-tree indexing (WPCB-Tree), and crash and data recovery on NAND flash and SSDs.

Its roots run deeper still: before flash, the lab worked on database systems and spatial/multimedia indexing — k-nearest-neighbor queries with semantic predicates (Spy-Tec+, RS-tree), distance browsing, and GIS object prefetching (1999–2011). Today the flash line feeds directly into the undergraduate File Processing course, whose assignments build an emulated NAND device driver and an FTL from scratch.

Architecture diagram of the WPCB-tree: a B+-tree module whose random writes pass through a transit buffer and write-pattern converter into sequential writes over the FTL and flash memory — WPCB-tree — a flash-aware B-tree over the FTL (Symmetry 2018)

연구 과제

Funded Projects연구 과제

National R&D and industry projects the lab runs or contributes to.

Click a column header to sort


Basic Research on Multimodal Korean Voice Phishing Detection: Adversarial Defense and Privacy Preservation in the LLM Era ^‡ 한국어 보이스피싱 다중모달 탐지의 기초연구: LLM 시대 적대적 방어와 프라이버시 보존을 중심으로 Role Principal-investigator proposal. Five pillars (Y1–Y5): cross-modal attention foundations; the KorCCVi-DF deepfake benchmark; certified defenses against LLM-paraphrase and acoustic attacks; large audio–language models with LoRA/PEFT; and federated learning with differential privacy and secure aggregation.	Ministry of Science and ICT (MSIT) 2026 기본연구 유형A (Basic Research, Type A) — 5-year proposal	—	—	Proposal
Development of Content Analysis and Type-Information Determination Technology for the Global Diffusion of Openly Licensed (Shared) Works 공유저작물의 글로벌 확산을 위한 콘텐츠 분석 및 유형정보 판단 기술 개발 Role Co-research (Soongsil University Industry–Academic Cooperation Foundation). Stage 1: metadata-learning auto-extraction (NER/BERT), a copyright-infringement-possibility AI algorithm, a multimodal (CLIP-based) survey, and a UCI-based metadata format. Stage 2: image–text multimodal AI analysis, large-scale training with standardized reliability enhancement, and open-platform pilot support. Demand organizations: Korea Copyright Commission (Gongu Madang) and Korea Culture Information Service Agency (KOGL).	Ministry of Culture, Sports and Tourism (MCST) · KOCCA 2025 문화체육관광 연구개발사업 (Culture, Sports and Tourism R&D Program)	2025.04 – 2026.12	Muhayu Inc. (lead) · HM Company · Soongsil University IACF	Active
SW-Centered University Program ^† SW중심대학 사업 Grant no. 2024-0-00071 Role Acknowledged as a funding source in the lab's copyright-metadata publications.	Ministry of Science and ICT (MSIT) · IITP SW중심대학 (Software-Centered University)	—	—	Active
MSIT/IITP research grant 2018-0-00209 (title not published) ^† 과학기술정보통신부/IITP 연구과제 2018-0-00209 (과제명 미공개) Grant no. 2018-0-00209 Role Acknowledged as a funding source of the Mathematics 2023 voice-phishing detection paper (Crossref funder data).	Ministry of Science and ICT (MSIT) · IITP	—	—	Completed

Listed from publication acknowledgements; project title and role pending confirmation with the funding agency.

Submitted proposal — under review, not a funded award.