Jaehyun Jeon

[E-mail]  [LinkedIn]  [GitHub]  [Google Scholar] 
prof_pic.jpg

👋 Hello! I’m Jaehyun, a Research Engineer at DialogTech Team in NC AI.

📍 Previously, I worked at Language Lab in LG AI Research. I received my M.S. (2026) and B.S. (2024, with Highest Honors) in Computer Science from Yonsei University, where I conducted research at MIR Lab (now SNU PI Lab) under supervision of Prof. Youngjae Yu, also mentored by Prof. Dae Hyun Kim

🤖 I am passionate about advancing multimodal AI systems that deliver real-world impact across diverse applications. My work focuses on developing and evaluating VLMs and LLMs, and building intent-aligned systems such as RAG or agent-based dialogue systems, while extending them to new domains such as UI/UX. Outside of work, I enjoy watching movies, taking photos, listening to music, and following sports.

Please feel free to reach out!

News

Apr 07, 2026 Our paper is accepted as a main conference paper at ACL 2026. See you in San Diego!
Jan 20, 2026 I am joining NC AI as a Research Engineer!
Sep 29, 2025 I am starting a research engineer internship at LG AI Research!
Aug 21, 2025 Our paper is accepted as a main conference paper at EMNLP 2025.

Publications

  1. ACL 2026 Main
    wiserUI-bench.png
    Do MLLMs Capture How Interfaces Guide User Behavior? A Benchmark for Multimodal UI/UX Design Understanding
    Jaehyun Jeon, Min Soo Kim, Jang Han Yoon, Sumin Shim, Yejin Choi, Hanbin Kim, Dae Hyun Kim, Youngjae Yu
    In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics, 2026
  2. A11YN.png
    A11YN: Aligning LLMs for Web Accessibility-Aware UI Generation
    Janghan Yoon, Jaegwan Cho, Junhyeok Kim, Jiwan Chung, Jaehyun Jeon, Seungwon Lim, Youngjae Yu
    arXiv preprint arXiv:2510.13914, 2026
  3. EMNLP 2025 Main
    premir.png
    Zero-shot Multimodal Document Retrieval via Cross-modal Question Generation
    Yejin Choi, Jaewoo Park, Janghan Yoon, Saejin Kim, Jaehyun Jeon, Youngjae Yu
    In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, 2025
  4. EMNLP 2024 Main
    visualpun.png
    Can Visual Language Models Resolve Textual Ambiguity with Visual Cues? Let Visual Puns Tell You!
    Jiwan Chung, Seungwon Lim, Jaehyun Jeon, Seungbeen Lee, Youngjae Yu
    In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, 2024

Selected Projects

2025

Lead
Automated Accommodation Image Assessment Based on User Preferences
Developed an end-to-end framework for assessing accommodation images based on human-perceived attractiveness

2024

Released a 70B LLaMA-based LLM full fine-tuned with curated data to be specialized in Korean
Link
Built RAG-based multimodal chatbot with webOS knowledge base, automating data preprocessing to answering pipeline

2023

Lead
User-interactive Image Captioning with Constrained Decoding
Developed an interactive image captioning system enabling users to fix specific words in the generated captions
Link
Lead
Visual Distractor: Vision-Language Models Fooled by Images
Proposed a benchmark measuring VLM hallucination vulnerabilities from visual interference in text question answering

2022

Lead
Establishment of Student Performance Indicators in Service
Identified students' actual math skill levels to highlight the value of the company's service
Link