Hello, I am Sumyeong Ahn, a Postdoctoral Fellow at MSU collaborating with Prof. Jiayu Zhou. I obtained my Ph.D. from KAIST under the guidance of Prof. Se-Young Yun (KAIST AI) and Yung Yi (KAIST EE). My research has centered on creating dependable AI solutions across diverse AI domains. In particular, my previous work delved into addressing issues like dataset bias, noisy labels, and class imbalances. Lately, I’ve embarked on a study of multi-modality, exploring various modalities such as Graph-Language, Visual-Language, and Sound-Language models.

---

Publications

C: Conference, W: Workshop, J: Journal, P: Preprint, M: Misc.
* : Equal Contribution ^ : Equal Advising

[P3] Prompt Learning with Noisy Labels (Modified for anonimity)
Sumyeong Ahn, Siqi Liang, Jiayu Zhou
Under Review.

[P2] ORBIS: Open Dataset Can Rescue You from Dataset Bias Problem
Sumyeong Ahn, Se-Young Yun
Under Review.

[P1] Large Language Models In Medical Term Classification And Unexpected Misalignment Between Response And Reasoning
Xiaodan Zhang, Sandeep Vemulapalli, Nabasmita Talukdar, Sumyeong Ahn, Jiankun Wang, Han Meng, Sardar Mehtab Bin Murtaza, Aakash Ajay Dave, Dmitry Leshchiner, Dimitri F. Joseph, Martin Witteveen-Lane, Dave Chesla, Jiayu Zhou, and Bin Chen
Under Review.

[C10] Fine-tuning Pre-trained Models for Robustness Under Noisy Labels
Sumyeong Ahn, Sihyeon Kim, Jongwoo Ko, Se-Young Yun
IJCAI’24 (To appear).

[C9] Active Prompt Learning in Vision Language Models
Jihwan Bang, Sumyeong Ahn, Jae-Gil Lee
CVPR’24 (To appear).

[C8] Large Language Models in Medical Term Classification and Unexpected Misalignment Between Response and Reasoning
Xiaodan Zhang, Sandeep Vemulapalli, Nabasmita Talukdar, Sumyeong Ahn, Jiankun Wang, Han Meng, Sardar Mehtab Bin Murtaza, Aakash Ajay Dave, Dmitry Leshchiner, Dimitri F Joseph, Martin Witteveen-Lane, Dave Chesla, Jiayu Zhou, Bin Chen
AMIA’24

[C7] NASH: A Simple Unified Framework of Structured Pruning for Accelerating Encoder-Decoder Language Models
Jongwoo Ko*, Seungjoon Park*, Yujin Kim, Sumyeong Ahn^, Du-Seong Chang, Euijai Ahn, Se-Young Yun^
EMNLP’23 (Findings)

[W3] Efficient Utilization of Pre-trained Model for Learning with Noisy Labels
Jongwoo Ko*, Sumyeong Ahn*, Se-Young Yun
Trustworthy ML Workshop at ICLR’23

[C6/W2] CUDA: Curriculum of Data Augmentation for Long-Tailed Recognition
Sumyeong Ahn*, Jongwoo Ko*, Se-Young Yun
ICLR’23 (Spotlight, notable-top-25%)
ML Safety Workshop at NeurIPS’22, NeurIPS’22 Workshop on Distribution Shifts: Connecting Methods and Applications
[Paper] [Code]

[C5/W1] Mitigating Dataset Bias by Using Per-sample Gradient
Sumyeong Ahn*, Seongyoon Kim*, Se-Young Yun
ICLR’23
ML Safety Workshop at NeurIPS’22, NeurIPS’22 Workshop on Distribution Shifts: Connecting Methods and Applications
[Paper] [Code]

[C4] Denoising After Entropy-based Debiasing A Robust Training Method for Dataset Bias with Noisy Labels
Sumyeong Ahn, Se-Young Yun
AAAI’23
[Paper] [Code]

[C3] Neuro-DCF: Design of Wireless MAC via Multi-Agent Reinforcement Learning Approach
Sangwoo Moon, Sumyeong Ahn Kyunghwan Son, Jinwoo Park, Yung Yi
ACM MobiHoc’21
[Paper]

[C2] Enlarging Discriminative Power by Adding an Extra Class in Unsupervised Domain Adaptation
Tran Hai H, Sumyeong Ahn, Taeyoung Lee, Yung Yi
ICPR’21
[Paper]

[J1] Multi-Armed Bandit with Additional Observations
Donggyu Yun, Sumyeong Ahn Alexandre Proutiere, Jinwoo Shin, Yung Yi
POMACS’18
[Paper]

[C1] Multi-Armed Bandit with Additional Observations
Donggyu Yun, Sumyeong Ahn Alexandre Proutiere, Jinwoo Shin, Yung Yi
SIGMETRICS’18
[Paper]

[M2] Client Sampling Algorithm in Federated Learning via Combinatorial Averaging and Multi-Armed Bandits
Sangmin Bae, Taehyeon Kim, Sumyeong Ahn, Sangmook Kim, Jongwoo Ko, Se-Young Yun
KIISE’22

[M1] An Implementation of Multi-Hop Voice Communication System using Drones
Yoonpyo Koo, Sumyeong Ahn, Kyounghwan Son, Suho Shin, Jeonghun Yu, Jaesin Kim, Yung Yi
KICS’17

---

Patent

[1] Method for Controlling Multi UAVs Based on Multi-Hop Wireless Mesh Networks
KOR patent number: 10-2019874-0000

---

Projects

Detailed information is summarized in the CV

Leading Projects

[5] Developing NAS algorithm for Layer-wise Communication Network Failure detection utilizing Multi-Layer datasets
KT (08/2021-08/2022)

[4] Application Service Implementation (Store Open/Close prediction) based on data from commercial stores’ power consumption
KEPRI (04/2018-12/2019)

[3] Versatile Network System Architecture for Multi-dimensional Diversity
IITP (04/2016-08/2019)

[2] Designing Matching Algorithm for Distributed Edge Devices
KISA (06/2015-06/2016)

[1] Modeling of Multi-Hop Wireless Networks for Military Applications
ADD (09/2015-12/2016)

Participating Student

[1] Video Transmission over Vehicular Networks
Bosch (01/2015-06/2015)

---

Education

[MSU], Postdoctoral Fellow in Computer Sicence and Engineering/ East Lansing, USA/ Sep 2023 - (Current)
[KAIST], Ph.D. in Graduate School of AI/ Seoul, South Korea/ July 2023
[KAIST], M.S. in Electrical Engineerin/ Daejeon, South Korea/ July 2017
[Korea Univ.], B.S. in School of Electrical Engineerin/ Seoul, South Korea/ Feb 2015

---

Reference

Prof. Jiayu Zhou: Professor at the Computer Science and Engineering, Michigan State University (MSU), USA, zhou [at] cse [dot] msu [dot] edu

Prof. Se-Young Yun: Professor at the Graduate School of AI, Korea Advanced Institute of Science and Technology (KAIST), South Korea, yunseyoung [at] kaist [dot] ac [dot] kr

Prof. Yung Yi: Professor at the Electrical Engineering, Korea Advance Institute of Science and Technology (KAIST), South Korea, yiyung [at] kaist [dot] edu