My work builds multimodal generative models that perceive and reason over audio, radio
frequency, and video signals, and makes them efficient enough to deploy
under real compute, bandwidth, and energy budgets. Representative topics:
-
Token efficiency in audio LLMs
(in submission)
-
Multimodal diffusion models with visual and physical signals
(CVPR Findings '26, one paper in submission)
-
AI-generated video detection and reasoning-based MLLMs
(ICLR '26)
-
Video streaming in LEO satellite networks
(INFOCOM '26)
-
Resource-efficient mobile and adaptive video systems
(CoNEXT '24, TMC '22, INFOCOM '19)
Acceptance rates are shown for conference papers where available. Full list on
Google Scholar.
Conference Papers
-
Diffusion2: Turning 3D Environments into Radio Frequency Heatmaps
Kyoungjun Park, Yifan Yang, Changhan Ge, Lili Qiu, Shiqi Jiang
-
VidGuard-R1: AI-Generated Video Detection and Explanation via Reasoning MLLMs and RL
Kyoungjun Park, Yifan Yang, Juheon Yi, Shicheng Zheng, Muhammad Muaz, Yifei Shen, Dongqi Han, Caihua Shan, Lili Qiu
-
Joint Optimization of Handoff and Video Rate in LEO Satellite Networks
Kyoungjun Park, Zhiyuan He, Cheng Luo, Yi Xu, Lili Qiu, Changhan Ge, Muhammad Muaz
-
Real-Time Neural Video Recovery and Enhancement on Mobile Devices
Zhaoyuan He, Yifan Yang, Lili Qiu, Kyoungjun Park, Yuqing Yang
-
EVSO: Environment-aware Video Streaming Optimization of Power Consumption
Kyoungjun Park, Myungchul Kim
Journal Articles
-
NeuSaver: Neural Adaptive Power Consumption Optimization for Mobile Video Streaming
Kyoungjun Park, Myungchul Kim, Laihyuk Park
-
Energy-Efficient Mobile Charging for Wireless Power Transfer in Internet of Things Networks
Woongsoo Na, Junho Park, Cheol Lee, Kyoungjun Park, Joongheon Kim, Sungrae Cho
Under Review
-
Importance-Aware Audio Token Compression in Frozen Audio LLMs
Kyoungjun Park, Lili Qiu
In submission
Preprint coming soon
-
World Models with Signals: Video Generation Conditioned on Wi-Fi, Radar, and Lidar
Kyoungjun Park, Shicheng Zheng, Yifan Yang, Junbo Li, Lili Qiu
-
Towards Practical Pain Monitoring: Ear-EEG with Foundation Models and Adaptive Inference
Muhammad Muaz, Yinan Zhu, Hyeongheon Cha, Haochen Zhao, Taeckyung Lee, Archit Patil, Kyoungjun Park, Taiting Lu, Gaoteng Zhao, Zilong Wang, Sung-Ju Lee, Lili Qiu
In submission
Preprint coming soon
-
MERGED: Multimodal Entity Resolution via Generated Expert Reasoning Distillation
You-Lin Chen, Kyoungjun Park, Prithviraj Sen, Bin Xu, Abhishek Prasad Ram Tripathi, Pedro Herrero-Vidal
In submission
Preprint coming soon
