Wonjun Kang
Scholar

Wonjun Kang

Google Scholar ID: Q-ARWkwAAAAJ
Seoul National University
Computer VisionDiffusion ModelsMultimodal Large Language Models
Citations & Impact
All-time
Citations
69
 
H-index
6
 
i10-index
3
 
Publications
15
 
Co-authors
30
list available
Resume (English only)
Academic Achievements
  • [{'PaperTitle': 'XQuant: Breaking the Memory Wall for LLM Inference with KV Cache Rematerialization', 'PublicationStatus': 'Preprint'}, {'PaperTitle': 'UNCAGE: Contrastive Attention Guidance for Masked Generative Transformers in Text-to-Image Generation', 'PublicationStatus': 'Preprint'}, {'PaperTitle': 'Draft-based Approximate Inference for LLMs', 'PublicationStatus': 'Preprint'}, {'PaperTitle': 'State-offset Tuning: State-based Parameter-Efficient Fine-Tuning for State Space Models', 'PublicationStatus': 'ACL 2025'}, {'PaperTitle': 'Development of OCR Service for Page-Level Recognition for Camera-Captured Document Images', 'PublicationStatus': 'IEEE Access'}, {'PaperTitle': 'VersaPRM: Multi-Domain Process Reward Model via Synthetic Reasoning Data', 'PublicationStatus': 'ICML 2025 (Oral, Top ~1%)'}, {'PaperTitle': 'TabFlex: Scaling Tabular Learning to Millions with Linear Attention', 'PublicationStatus': 'ICML 2025 (Spotlight, Top 2.6%)'}, {'PaperTitle': 'Parameter-Efficient Fine-Tuning of State Space Models', 'PublicationStatus': 'ICML 2025'}, {'PaperTitle': 'In-batch Ensemble Drafting: Robust Speculative Decoding for LVLMs', 'PublicationStatus': 'ICLR 2025 Workshop'}, {'PaperTitle': 'Counting Guidance for High Fidelity Text-to-Image Synthesis', 'PublicationStatus': 'WACV 2025 (Oral)'}, {'PaperTitle': 'Can MLLMs Perform Text-to-Image In-Context Learning?', 'PublicationStatus': 'COLM 2024'}, {'PaperTitle': 'Eta Inversion: Designing an Optimal Eta Function for Diffusion-based Real Image Editing', 'PublicationStatus': 'ECCV 2024'}, {'PaperTitle': 'Face Swapping for Low-Resolution and Occluded Images In-the-Wild', 'PublicationStatus': 'IEEE Access'}, {'PaperTitle': 'One-Shot Face Reenactment on Megapixels', 'PublicationStatus': 'Preprint'}]
Background
  • PhD student in the Department of Electrical and Computer Engineering at Seoul National University, advised by Prof. Nam Ik Cho. Also works as an AI research engineer at the Korean startup FuriosaAI.