Daewon Chae

Ph.D. Student, Electrical and Computer Engineering
University of Michigan

Hello! I am a second-year Ph.D. student at the University of Michigan, advised by Prof. Hun-Seok Kim. My current research interests are efficient ML and model compression.

Previously, I received my M.S. in Computer Science from Korea University under Prof. Jinkyu Kim. I also spent time as a visiting student at the University of Wisconsin–Madison, working with Prof. Kangwook Lee.

Daewon Chae

News

  1. Our paper got accepted to NeurIPS 2026
  2. Our paper got accepted to TMLR 2025 (J2C Certification)
  3. Started my Ph.D. journey at the University of Michigan
  4. Our paper got accepted to ICML 2025 (Oral)
  5. Our paper got accepted to AAAI 2025 (Oral)

Publications

* equal contribution  ·  † co-corresponding author

2026

  1. DyRA: Dynamic Residual Approximation for Efficient Matrix Multiplication in DNNs

    Daewon Chae, Hyunwon Chung, Changwoo Lee, Hun-Seok Kim

    Neural Information Processing Systems (NeurIPS), 2026

    paper code

  2. Streaming Dense Voxel Representations for 3D Occupancy Prediction

    Seokha Moon, Janghyun Baek, Yujin Jeong, Daewon Chae, Giseop Kim, Jungbeom Lee, Jinkyu Kim, Sunwook Choi

    European Conference on Computer Vision (ECCV), 2026

    paper code

  3. AxisGuide: Grounding Robot Action Coordinate System in RGB Observations for Robust Visuomotor Manipulation

    Jiyun Jang, Yujin Sung, Woosung Joung, Daewon Chae, Sangwon Lee, Sohwi Kim, Jinkyu Kim, Jungbeom Lee

    Robotics: Science and Systems (RSS), 2026

    paper code

  4. Scene Graph-Guided Proactive Replanning for Failure-Resilient Embodied Agent

    Che Rin Yu, Daewon Chae, Dabin Seo, Sangwon Lee, Hyeongwoo Im, Jinkyu Kim

    Autonomous Agents and Multiagent Systems (AAMAS), 2026

    RSS Workshop on Semantic Reasoning and Goal Understanding in Robotics (RSSW), 2025

    paper

2025

  1. ENTP: Encoder-only Next Token Prediction

    Ethan Ewer*, Daewon Chae*, Thomas Zeng*, Jinkyu Kim, Kangwook Lee

    Transactions on Machine Learning Research (TMLR), 2025 J2C Certification

    NeurIPS Workshop on What Can('t) Transformers Do? (NeurIPSW), 2025 Spotlight

    paper code summary

  2. Hard-Negative Prototype-Based Regularization for Few-Shot Class-Incremental Learning

    Seongbeom Park, Hyunju Yun, Daewon Chae, Sungyoon Kim, Suhong Moon, Minwoo Kang, Seunghyun Park, Jinkyu Kim

    Transactions on Machine Learning Research (TMLR), 2025

    paper

  3. SemanticControl: A Training-Free Approach for Handling Loosely Aligned Visual Conditions in ControlNet

    Woosung Joung*, Daewon Chae*, Jinkyu Kim

    British Machine Vision Conference (BMVC), 2025

    paper project page

  4. VersaPRM: Multi-Domain Process Reward Model via Synthetic Reasoning Data

    Thomas Zeng, Shuibai Zhang, Shutong Wu, Christian Classen, Daewon Chae, Ethan Ewer, …, Minjae Lee, Heeju Kim, Wonjun Kang, Jackson Kunde, Ying Fan, Jungtaek Kim, Hyung Il Koo, Kannan Ramchandran, Dimitris Papailiopoulos, Kangwook Lee

    International Conference on Machine Learning (ICML), 2025 Oral

    paper code summary

  5. DiffExp: Efficient Exploration in Reward Fine-tuning for Text-to-Image Diffusion Models

    Daewon Chae*, June Suk Choi*, Jinkyu Kim†, Kimin Lee†

    AAAI Conference on Artificial Intelligence (AAAI), 2025 Oral

    paper

2024

  1. InstructBooth: Instruction-following Personalized Text-to-Image Generation

    Daewon Chae, Nokyung Park, Jinkyu Kim, Kimin Lee

    ICML Workshop on Foundation Models in the Wild (ICMLW), 2024

    paper project page

  2. Clustering-based Image-Text Graph Matching for Domain Generalization

    Nokyung Park, Daewon Chae, Jeongyong Shim, Sangpil Kim, Eun-Sol Kim, Jinkyu Kim

    International Conference on Pattern Recognition (ICPR), 2024

    paper

  3. Text-driven Prototype Learning for Few-Shot Class-Incremental Learning

    Seongbeom Park*, Haeji Jung*, Daewon Chae, Hyunju Yun, Sungyoon Kim, Suhong Moon, Seunghyun Park, Jinkyu Kim

    International Conference on Pattern Recognition (ICPR), 2024

    paper

2023

  1. Re-ID Technology for V2I based Cooperative Driving Protocol

    Junhyek Jang, Kitaeg Lim, Sanghun Yoon, Daewon Chae, Soohyun Jang

    International Conference on Ubiquitous and Future Networks (ICUFN), 2023

    paper