← Back to site
Habin Lim

Habin Lim

Generative AI Researcher · M.S.–Ph.D. Student
VGI Lab, Korea University

Summary

I am an M.S.–Ph.D. student at the Visual & Geometric Intelligence (VGI) Lab, Korea University. My research builds on diffusion-based generative models and spans a broad range of multimodal problems — image, video, audio, 3D, and LLMs — with the goal of contributing to both theoretical understanding and practical applications. Outside the lab, I love performing as the lead vocalist in a rock band.

Publications

  1. Diffusion-Based AI System for Second Language Reading: Development, Text Quality Validation, and Learner Perception

    H. Lim*, Euijune Lee*, Si Eun Kwoun, Sung-Ho Bae, G.-M. Park, Yohan Hwang, and Sangmin-Michelle Lee†

    Computers & Education, vol. 251, 105653, Oct. 2026 (JCR: 0.6%, IF: 10.8) · link

  2. OSVE: One Step Video Editing with One Step Diffusion Models

    H. Lim and G.-M. Park†

    European Conference on Computer Vision (ECCV), Malmö, Sweden, Sep. 2026 · link TBD

  3. ConceptSplit: Decoupled Multi-Concept Personalization of Diffusion Models via Token-wise Adaptation and Attention Disentanglement

    H. Lim, Youngseob Won, J. Seo, and G.-M. Park†

    International Conference on Computer Vision (ICCV), Hawaii, U.S.A., Oct. 2025 · link

  4. Good-Text Inversion for Faster Real Image Editing using Guided Diffusion Models

    H. Lim and G.-M. Park

    Korea Software Congress (KSC), Yeosu, Republic of Korea, Dec. 2024 · link

  5. LFS-DIFFUSION: Lifelong Few-shot Image Generation In Latent Diffusion Model Via LeFT Weight Modulator

    H. Lim, Y.-S. Won, SeonHyeok Kim, Yeonjin Seo, and G.-M. Park

    Korea Computer Congress (KCC), Jeju Island, Republic of Korea, Jun. 2024 [Award] · link

Experience

Education

Honors & Awards

Skills

Artificial Intelligence · Computer Vision · Generative Models · Diffusion Models · Python · PyTorch