Scholar
Jeonghun Baek
Google Scholar ID: Bl5zbmUAAAAJ
The University of Tokyo
Citations & Impact
Citations
1,264
H-index
7
i10-index
6
Publications
18
Co-authors
26
list available
Publications
Resume (English only)
Academic Achievements
- Published 'LLM-Based Explainable Detection of LLM-Generated Code in Python Programming Courses' at ACM SIGCSE 2026
- Released 'MangaVQA and MangaLMM: A Benchmark and Specialized Model for Multimodal Manga Understanding' on arXiv 2025; accepted as oral presentation at ICCV COMIQ Workshop 2025
- Published multiple papers at ICCV 2025 workshops on topics including LMM-as-a-Judge, image harmonization evaluation, and safety judgment via text-to-image generation
- Published 'Harnessing PDF Data for Improving Japanese Large Multimodal Models' in ACL Findings 2025
- Co-developed JMMMU, a Japanese Massive Multi-discipline Multimodal Understanding Benchmark; presented at NAACL 2025 and NeurIPS EvalEval Workshop 2024 (oral)
- Presented poster 'Leveraging LLM for Detecting and Explaining LLM-generated Code in Python Programming Courses' at SIGCSE 2025
- Published 'Cross-Lingual Learning in Multilingual Scene Text Recognition' at ICASSP 2024
- Published work on character image combination for synthetic data in ICCV Workshops 2023
- Published 'COO: Comic Onomatopoeia Dataset for Recognizing Arbitrary or Truncated Texts' at ECCV 2022
- Published 'What If We Only Use Real Datasets for Scene Text Recognition? Toward Scene Text Recognition With Fewer Labels' at CVPR 2021
- Co-authored 'Character Region Attention For Text Spotting' at ECCV 2020
