Seohyeon Cha

Seohyeon Cha

I am a Ph.D. student in Electrical and Computer Engineering at The University of Texas at Austin, advised by Prof. Haris Vikalo. I earned my M.S. and B.S. in Electrical Engineering at KAIST, with Summa Cum Laude honors for my B.S.

I develop resource-efficient machine learning methods for deploying and adapting AI models across distributed systems and devices with limited hardware resources. My research aims to reduce the memory, computation, and communication costs of training and inference. Recent work includes LLM post-training quantization, federated and continual learning, and resource-aware model placement across cloud, edge, and embedded systems.

Summer 2027: I am actively seeking research internship opportunities.

News

Sep 2026Our paper CoreQ was accepted to NeurIPS 2026 as a Spotlight (top 5.1% of accepted papers).
Summer 2026I started a research internship with the embedded AI team at Texas Instruments Kilby Labs.
Jun 2026Our paper FedProTIP was accepted to Transactions on Machine Learning Research.
Apr 2026Our paper FedRot-LoRA was accepted to ICML 2026.
Jan 2026Our paper Quantized Gradient Projection for Memory-Efficient Continual Learning was accepted to ICLR 2026.

Education

2024 - PresentThe University of Texas at Austin, Austin, TX
Ph.D. in Electrical and Computer Engineering
Advisor: Prof. Haris Vikalo
2022 - 2024KAIST, Daejeon, South Korea
M.S. in Electrical Engineering
Advisor: Prof. Joonhyuk Kang
2017 - 2022KAIST, Daejeon, South Korea
B.S. in Electrical Engineering
Summa Cum Laude

Experience

Summer 2026Texas Instruments Kilby Labs
Research Intern, Embedded AI Team
2024 - 2025The University of Texas at Austin
Research Assistant
Summer 2019SK hynix
Undergraduate Summer Intern

Publications

* indicates equal contribution.

  1. PTQ regularized calibration
    CoreQ: Learning-Free Mismatch Correction and Successive Rounding for Quantization
    Seohyeon Cha, Huancheng Chen, Dongjun Kim, Haoran Zhang, Kevin Chan, Gustavo de Veciana, Haris Vikalo
    Advances in Neural Information Processing Systems (NeurIPS), 2026 · Spotlight (top 5.1% of accepted papers)
    arXiv
  2. multi-layer hierarchical inference
    Online Learning for Multi-Layer Hierarchical Inference under Partial and Policy-Dependent Feedback
    Haoran Zhang, Seohyeon Cha, H. Burak Beytur, Kevin S. Chan, Gustavo de Veciana, Haris Vikalo
    arXiv preprint, 2026
    arXiv
  3. joint model onloading and offloading
    Batching-Aware Joint Model Onloading and Offloading for Hierarchical Multi-Task Inference
    Seohyeon Cha, Kevin Chan, Gustavo de Veciana, Haris Vikalo
    arXiv preprint, 2025
    arXiv
  4. task-agnostic continual federated learning
    FedProTIP: Task-Agnostic Federated Continual Learning via Replay-Free Gradient Projection
    Seohyeon Cha*, Huancheng Chen*, Haris Vikalo
    Transactions on Machine Learning Research, 2026
    arXiv
  5. GeFL system model
    GeFL: Model-Agnostic Federated Learning with Generative Models
    Honggu Kang*, Seohyeon Cha*, Jiwan Seo, Joonhyuk Kang
    IEEE Transactions on Mobile Computing, 2025
    paperproject page
  6. NeFL nested scaling
    NeFL: Nested Model Scaling for Federated Learning with System Heterogeneous Clients
    Honggu Kang, Seohyeon Cha, Jinwoo Shin, Jongmyeong Lee, Joonhyuk Kang
    IEEE Transactions on Mobile Computing, 2025
    arXivproject page
  7. FedRot-LoRA rotational alignment
    FedRot-LoRA: Mitigating Rotational Misalignment in Federated LoRA
    Haoran Zhang, Dongjun Kim, Seohyeon Cha, Haris Vikalo
    International Conference on Machine Learning (ICML), 2026
    arXiv
  8. Quantized Gradient Projection Memory
    Quantized Gradient Projection for Memory-Efficient Continual Learning
    Dongjun Kim, Seohyeon Cha, Huancheng Chen, Chao Wang, Haris Vikalo
    International Conference on Learning Representations (ICLR), 2026
    OpenReview
  9. Bayesian GNN conformal prediction
    On the Temperature of Bayesian Graph Neural Networks for Conformal Prediction
    Seohyeon Cha, Honggu Kang, Joonhyuk Kang
    NeurIPS 2023 GLFrontiers Workshop
    paper
  10. intelligent transmitting surface
    Intelligent Surface-aided Transmit-array Antenna in mmWave Communication System with Historical Channel Observation
    Seohyeon Cha, Sanghyuk Kim, Jiwan Seo, Joonhyuk Kang
    IEEE ICCE-Asia, 2022
    paper

Honors and Teaching

Honors and Awards
  • Best Project Award, ML on Real World Networks, UT Austin (2024)
  • Korean Governmental Scholarship, KAIST Graduate (2022 - 2024)
  • National Science and Engineering Scholarship, Academic Excellence (2019 - 2021)
  • Korean Governmental Scholarship, KAIST Undergraduate (2017 - 2018)
Teaching
  • Undergraduate Individual Study Assistant, KAIST (2023)
  • Teaching Assistant, KAIST (2022 - 2023)
  • Freshman Tutoring, KAIST (2018 - 2019)