Member Profile
Education
- 2022.03 - 2027.08 B.S. in Computer Engineering, Hankyong National University, South Korea
Publications
Published (9)
-
[Journal] AgentReport: A Multi-Agent LLM Approach for Automated and Reproducible Bug Report Generation
In Applied Sciences, Vol. 15, No. 22, pp. 1-31, 2025
-
[Journal] (KCC2025 우수논문) QLoRA와 CTQRS 구조화 프롬프트, CoT·Few-shot 전략을 결합한 오픈 소스 버그 리포트 자동 생성 기법
정보과학회논문지 : 소프트웨어 및 응용, Vol. 53, No. 3, pp. 217-229, 2026
-
[Conference] Fault Exposure as a Validity Endpoint for LLM-Generated Unit Tests: Six LLMs Across Eight Defects4J Projects
In Proc. of the 33rd Asia-Pacific Software Engineering Conference (APSEC 2026), Bali, Indonesia, to appear
-
[Conference] 생성형 LLM 기반 SATD 평가에서 기본 임계값 0.5의 실패: 점수 압축과 조건부 회복 분석
한국컴퓨터종합학술대회 (KCC), 2026
-
[Conference] 실행 컨텍스트 정합성에 기반한 LLM 결함 위치 추정 결과의 안정화 기법
한국소프트웨어공학학술대회 (KCSE), 2026
-
[Conference] 검색 전략이 LLM 기반 버그 리포트 자동 생성 성능에 미치는 영향 분석
한국소프트웨어공학학술대회 (KCSE), 2026
-
[Conference] 설명 가능한 인공지능(XAI)과 사고연쇄추론(CoT)에 기반한 신뢰성 높은 대규모 언어모델 버그 리포트 생성 기법
한국소프트웨어종합학술대회 (KSC), 2025
-
[Conference] 유사 리포트 문맥 보강을 활용한 Retrieval-Augmented LLM 기반 버그 리포트 생성 기법
한국컴퓨터종합학술대회 (KCC), 2025
-
2025 한국컴퓨터종합학술대회 (KCC 2025) 우수논문상
한국정보과학회장상
Under Review (3)
-
[Journal] Ranking Sensitivity to Automatic Evaluation Specifications: LLMs versus Retrieval-Copy Baselines for Bug-Report Section Completion
under review at the Empirical Software Engineering (EMSE)
-
[Conference] Title Withheld for Double-Blind Review
under review at the 49th IEEE/ACM International Conference on Software Engineering (ICSE 2027)
-
[Conference] Title Withheld for Double-Blind Review
under review at the 49th IEEE/ACM International Conference on Software Engineering (ICSE 2027)
Planned (4)
-
[Journal] Patch Summaries Change Approval and Response Validity in Local Code Review
Planned submission to Empirical Software Engineering (EMSE)
-
[Journal] Are Software Assurance Gates Complementary? Conditional Detection of Known Defects in AI-Generated Changes
Planned submission to Information and Software Technology (IST)
-
[Journal] Do Coding-Agent Findings Survive Model Generations? A Longitudinal Reproduction Study
Planned submission to Empirical Software Engineering (EMSE)
-
[Journal] Hallucination, Context Faithfulness, and Template Fidelity in Bug Report Generation with Open-Weight Large Language Models: Comparing Prompting, Retrieval Augmentation, and Direct Preference Optimization
Planned submission to IEEE Transactions on Software Engineering (TSE)