Scholar
Jaehyeong Jo
Google Scholar ID: sB2n9ncAAAAJ
PhD Student, KAIST
Citations & Impact
Citations
689
H-index
7
i10-index
7
Publications
13
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- Publications:
- - Frame Guidance: Training-Free Guidance for Frame-Level Control in Video Diffusion Models (Preprint, 2025)
- - Continuous Diffusion Model for Language Modeling (NeurIPS, 2025)
- - Silent Branding Attack: Trigger-free Data Poisoning Attack on Text-to-Image Diffusion Models (CVPR, 2025)
- - Conditional Synthesis of 3D Molecules with Time Correction Sampler (NeurIPS, 2024)
- - Identity Decoupling for Multi-Subject Personalization of Text-to-Image Models (NeurIPS, 2024)
- - Generative Modeling on Manifolds Through Mixture of Riemannian Diffusion Processes (ICML, 2024)
- - Graph Generation with Diffusion Mixture (ICML, 2024)
- - DiffusionNAG: Task-guided Neural Architecture Generation with Diffusion Models (ICLR, 2024)
- - Text-Conditioned Sampling Framework for Text-to-Image Generation with Masked Generative Models (ICCV, 2023)
- - Exploring Chemical Space with Score-based Out-of-distribution Generation (ICML, 2023)
- - Score-based Generative Modeling of Graphs via the System of Stochastic Differential Equations (ICML, 2022)
- - Edge Representation Learning with Hypergraphs (NeurIPS, 2021)
- - Antibody-SGM: Antigen-Specific Joint Design of Antibody Sequence and Structure Using Diffusion Models (ICML 2023 Workshop on Computational Biology)
Research Experience
- Former Research Intern at Meta.
Education
- PhD candidate at KAIST AI, advised by Prof. Sung Ju Hwang.
Background
- Research Interests: Exploring the physical world through the lens of geometry. Developed diffusion models that incorporate geometric principles to generate structured data such as graphs and data on Riemannian manifolds, applied to drug discovery, language modeling, and neural architecture search.
Miscellany
- Contact: Twitter: JaehyeongJo, Email: harryjo97 [at] kaist.ac.kr
Co-authors