Thanh-Tung Nguyen

About Me

I am a Ph.D. student at Collaborative Distributed Systems & Networks (CDSN) Lab, Korea Advanced Institute of Science and Technology (KAIST) under the advisorship of Prof. Dongman Lee. Previously, I was awarded the Korean Government Scholarship (KGSP) to pursue my Masterโ€™s degree in Information Engineering at Chungbuk National University, where I was advised by Prof. Taehong Kim.

My research focuses on the design and development of computing systems, particularly at the edge of the network. I am especially interested in areas such as edge AI and on-device AI, where inference and training can be performed on tiny, resource-constrained devices. I believe, in the future, these systems will become the backbone of a wide range of applications, including video analytics, IoT sensing, and smart mobility.

In my free time, I enjoy staying active. Whether itโ€™s running, hiking, playing football, or practicing calisthenics, I can never get enough of it.

Example Talk
Example Talk

An example talk using Hugo Blox Builder's Markdown slides feature.

Jun 1, 2030

FCPO: Federated Continual Policy Optimization for Real-Time High-Throughput Edge Video Analytics

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Duis posuere tellus ac convallis placerat. Proin tincidunt magna sed ex sollicitudin condimentum.

Jul 25, 2025

FCPO: Federated Continual Policy Optimization for Real-Time High-Throughput Edge Video Analytics
FCPO: Federated Continual Policy Optimization for Real-Time High-Throughput Edge Video Analytics

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Duis posuere tellus ac convallis placerat. Proin tincidunt magna sed ex sollicitudin condimentum.

Jul 25, 2025

How2Compress: Scalable and Efficient Edge Video Analytics via Adaptive Granular Video Compression

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Duis posuere tellus ac convallis placerat. Proin tincidunt magna sed ex sollicitudin condimentum.

Jul 21, 2025

๐ŸŽ‰ Our paper has been accepted at ACM Multimedia 2025 ๐ŸŽ‰

Our paper titled "*How2Compress: Scalable and Efficient Edge Video Analytics via Adaptive Granular Video Compression*" has been accepted to ACM Multimedia 2025, the very top conference in the domain.

Jul 10, 2025

OCTOPINF: Workload-Aware Inference Serving for Edge Video Analytics
OCTOPINF: Workload-Aware Inference Serving for Edge Video Analytics

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Duis posuere tellus ac convallis placerat. Proin tincidunt magna sed ex sollicitudin condimentum.

Mar 21, 2025

OCTOPINF: Workload-Aware Inference Serving for Edge Video Analytics
OCTOPINF: Workload-Aware Inference Serving for Edge Video Analytics

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Duis posuere tellus ac convallis placerat. Proin tincidunt magna sed ex sollicitudin condimentum.

Mar 21, 2025

๐ŸŽ‰ Our paper has been accepted at PerCom 2025 ๐ŸŽ‰

Our paper titled "*OCTOPINF: Workload-Aware Inference Serving for Edge Video Analytics*" has been accepted to PerCom 2025. The conference remains competitive with only 15-17% acceptance rate.

Dec 20, 2024

MetaWeather: Few-Shot Weather-Degraded Image Restoration
MetaWeather: Few-Shot Weather-Degraded Image Restoration

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Duis posuere tellus ac convallis placerat. Proin tincidunt magna sed ex sollicitudin condimentum.

Oct 4, 2024

MetaWeather: Few-Shot Weather-Degraded Image Restoration
MetaWeather: Few-Shot Weather-Degraded Image Restoration

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Duis posuere tellus ac convallis placerat. Proin tincidunt magna sed ex sollicitudin condimentum.

Oct 4, 2024