Yongjian Fu

I am an Assistant Researcher and Shuimu Scholar in the Department of Computer Science and Technology at Tsinghua University, working with Prof. Ju Ren. Prior to that, I received my B.S. and Ph.D. degrees in Computer Science from Central South University in 2021 and 2025, respectively, advised by Prof. Ju Ren. I worked closely with Prof. Lili Qiu and Prof. Hao Pan at MSRA from 2022 to 2024, and with Prof. Xinyu Zhang at UCSD as a visiting scholar from 2024 to 2025.

[Recruitment] We are looking for research interns to work on research projects in edge intelligence, embodied AI, and LLM agent.

News

Sep 25, 2026	TIMA and Argus have been accepted by NeurIPS’26 and SenSys’27, respectively!
Sep 23, 2026	Honored to receive the ACM SIGMOBILE China Doctoral Dissertation Award!
Aug 05, 2026	Invited to serve as TPC for MobiCom’27 , please consider submitting!
Jun 30, 2026	Our works Prism, XBandGS and MetaX have been accepted by MobiCom 2026.
Jun 25, 2026	Our work WDNN received the 2026 WAIC Young Outstanding Paper Award (世界人工智能大会青年优秀论文奖).

Selected publications

NeurIPS’26

TIMA: Test-Time Internalization for Agentic Memory

Xiaohang Sui , Yongjian Fu*, Yizhe Zhao , and 2 more authors

In Advances in Neural Information Processing Systems , 2026

(* Corresponding author)

Bib

@inproceedings{sui2026tima,
  title = {{TIMA}: Test-Time Internalization for Agentic Memory},
  author = {Sui, Xiaohang and Fu*, Yongjian and Zhao, Yizhe and Yue, Sheng and Ren, Ju},
  booktitle = {Advances in Neural Information Processing Systems},
  year = {2026},
  note = {(* Corresponding author)},
}

SenSys’27

Argus: Self-Scanning Programmable Metasurfaces for Multi-Person Context, Respiration, and Beyond

Linghui Zhong , Liyao Li , Shuning Wang , and 5 more authors

In The 25th ACM Conference on Embedded Networked Sensor Systems , 2027

(* Corresponding author)

Bib

@inproceedings{zhong2027argus,
  title = {Argus: Self-Scanning Programmable Metasurfaces for Multi-Person Context, Respiration, and Beyond},
  author = {Zhong, Linghui and Li, Liyao and Wang, Shuning and Fu*, Yongjian and Pan, Hao and Zhang, Yongmin and Ren, Ju and Xie, Yaxiong},
  booktitle = {The 25th ACM Conference on Embedded Networked Sensor Systems},
  year = {2027},
  note = {(* Corresponding author)},
}

MobiCom’26

Prism: Efficient Zeroth-Order Fine-Tuning for On-Device LLMs

Yongjian Fu, Wenyu Qi , Yizhe Zhao , and 5 more authors

In Proceedings of the 32nd Annual International Conference on Mobile Computing and Networking , Nov 2026

Bib

@inproceedings{MobiCom26_Prism,
  author = {Fu, Yongjian and Qi, Wenyu and Zhao, Yizhe and Sui, Xiaohang and Lu, Jun and Deng, Yongheng and Pan, Hao and Ren, Ju},
  title = {Prism: Efficient Zeroth-Order Fine-Tuning for On-Device LLMs},
  booktitle = {Proceedings of the 32nd Annual International Conference on Mobile Computing and Networking},
  series = {ACM MobiCom},
  year = {2026},
  address = {Austin, USA},
  month = nov,
}

MobiCom’26

Cross-Band RF Radiance Field Modeling via 3D Gaussian Splatting and Propagation Mixture-of-Experts

Zechen Li , Hao Pan , Yu Lu , and 3 more authors

In Proceedings of the 32nd Annual International Conference on Mobile Computing and Networking , Nov 2026

Bib

@inproceedings{MobiCom26_XBandGS,
  author = {Li, Zechen and Pan, Hao and Lu, Yu and Wang, Yezhou and Fu, Yongjian and Xue, Guangtao},
  title = {Cross-Band RF Radiance Field Modeling via 3D Gaussian Splatting and Propagation Mixture-of-Experts},
  booktitle = {Proceedings of the 32nd Annual International Conference on Mobile Computing and Networking},
  series = {ACM MobiCom},
  year = {2026},
  address = {Austin, USA},
  month = nov,
}

MobiCom’26

Mechanically Programmable Reflective Metasurfaces: A General Low‑Cost Paradigm for High‑Frequency Wavefront Control

Yezhou Wang , Xijun Zhang , Hao Pan , and 4 more authors

In Proceedings of the 32nd Annual International Conference on Mobile Computing and Networking , Nov 2026

Bib

@inproceedings{MobiCom26_MetaX,
  author = {Wang, Yezhou and Zhang, Xijun and Pan, Hao and Fu, Yongjian and Zhao, Yimin and Kong, Linghe and Xue, Guangtao},
  title = {Mechanically Programmable Reflective Metasurfaces: A General Low‑Cost Paradigm for High‑Frequency Wavefront Control},
  booktitle = {Proceedings of the 32nd Annual International Conference on Mobile Computing and Networking},
  series = {ACM MobiCom},
  year = {2026},
  address = {Austin, USA},
  month = nov,
}

MobiCom’26

Unlocking Practical Cardiac Monitoring Capabilities on True Wireless Stereo Earbuds

Yongjian Fu, Ke Sun , Xinyu Zhang , and 5 more authors

In Proceedings of the 32nd Annual International Conference on Mobile Computing and Networking , Nov 2026

Bib

@inproceedings{MobiCom26_TWSEarbuds,
  author = {Fu, Yongjian and Sun, Ke and Zhang, Xinyu and Wang, Ruyao and Li, Xinyi and Pan, Hao and Zhang, Yaoxue and Ren, Ju},
  title = {Unlocking Practical Cardiac Monitoring Capabilities on True Wireless Stereo Earbuds},
  booktitle = {Proceedings of the 32nd Annual International Conference on Mobile Computing and Networking},
  series = {ACM MobiCom},
  year = {2026},
  address = {Austin, USA},
  month = nov,
}

MobiCom’26

MANA: Towards Efficient Mobile Ad Detection via Multimodal Agentic UI Navigation

Yizhe Zhao , Yongjian Fu*, Zihao Feng , and 4 more authors

In Proceedings of the 32nd Annual International Conference on Mobile Computing and Networking , Nov 2026

(* Corresponding author)

Bib

@inproceedings{MobiCom26_MANA,
  author = {Zhao, Yizhe and Fu*, Yongjian and Feng, Zihao and Pan, Hao and Deng, Yongheng and Zhang, Yaoxue and Ren, Ju},
  title = {MANA: Towards Efficient Mobile Ad Detection via Multimodal Agentic UI Navigation},
  booktitle = {Proceedings of the 32nd Annual International Conference on Mobile Computing and Networking},
  series = {ACM MobiCom},
  year = {2026},
  address = {Austin, USA},
  month = nov,
  note = {(* Corresponding author)},
}

MobiCom’26

Attacking mmWave Imaging Using Metasurfaces Based on Intensity and Time-of-Flight Manipulation

Shuning Wang , Hao Pan , Yongjian Fu*, and 7 more authors

In Proceedings of the 32nd Annual International Conference on Mobile Computing and Networking , Nov 2026

(* Corresponding author)

Bib

@inproceedings{MobiCom26_MMWaveAttack,
  author = {Wang, Shuning and Pan, Hao and Fu*, Yongjian and Li, Xinyi and Zhong, Linghui and Xie, Yaxiong and Sun, Ke and Xue, Guangtao and Zhang, Yaoxue and Ren, Ju},
  title = {Attacking mmWave Imaging Using Metasurfaces Based on Intensity and Time-of-Flight Manipulation},
  booktitle = {Proceedings of the 32nd Annual International Conference on Mobile Computing and Networking},
  series = {ACM MobiCom},
  year = {2026},
  address = {Austin, USA},
  month = nov,
  note = {(* Corresponding author)},
}

MobiCom’25
WDNN: Weighted Diffractive Neural Network for Physical-layer RF Signal Processing

Yezhou Wang^ , Yongjian Fu^, Hao Pan , and 5 more authors

In Proceedings of the 31st Annual International Conference on Mobile Computing and Networking , Nov 2025

(^ Co-primary authors)

Abs Bib

Diffractive neural networks (NNs) have garnered attention for directly implementing wireless signal processing at the physical layer. However, they are limited by a constrained weight learning space and activation functions, which restricts their data processing capabilities. To address this, we propose an RF circuit-based weighted diffraction NN (WDNN) that rivals digital NNs in processing ability. We design a weighted asymmetric RF coupler unit that, when stacked into a network, enables diffractive propagation with arbitrary connection weights. Additionally, an activation module is introduced that utilizes RF amplifiers operating in their nonlinear regions. We validate the effectiveness of the proposed WDNN through three tasks: 32-level amplitude modulated (AM) signal decoding, 31-class angle of arrival (AoA) estimation, and 2-class Wi-Fi based fall detection. After training, WDNN achieves the accuracy of 98.5%, 93.7%, and 90.8% in the AM decoding, AoA estimation, and fall detection tasks, respectively; while the diffractive NN SOTA achieves only 21.6%, 16.9%, and 63.3%. We also implement the prototypes of WDNN and SOTA, and real-world experimental results demonstrate that our method achieves an average accuracy improvement of up to 76.85% across various tasks compared to SOTA.
@inproceedings{WDNN_yezhou, author = {Wang^, Yezhou and Fu^, Yongjian and Pan, Hao and Hu, Qinyun and Qiu, Lili and Chen, Yi-Chao and Xue, Guangtao and Ren, Ju}, title = {WDNN: Weighted Diffractive Neural Network for Physical-layer RF Signal Processing}, year = {2025}, isbn = {9798400711299}, publisher = {Association for Computing Machinery}, address = {New York, NY, USA}, url = {https://doi.org/10.1145/3680207.3765240}, doi = {10.1145/3680207.3765240}, booktitle = {Proceedings of the 31st Annual International Conference on Mobile Computing and Networking}, pages = {697–711}, numpages = {15}, keywords = {diffractive neural network, signal processing, radio-frequency hardware}, location = {Kerry Hotel, Hong Kong, Hong Kong, China}, series = {ACM MOBICOM '25}, month = nov, note = {(^ Co-primary authors)}, }

MobiSys’25

Towards Distance-Adaptive Wireless Charging

Lili Chen , Yizhe Zhao , Shuning Wang , and 5 more authors

In Proceedings of the 23rd Annual International Conference on Mobile Systems, Applications and Services , Jun 2025

Bib

@inproceedings{MobiSys25_DistanceAdaptiveCharging,
  author = {Chen, Lili and Zhao, Yizhe and Wang, Shuning and Zhong, Linghui and Fu, Yongjian and Yue, Sheng and Ren, Ju and Zhang, Yaoxue},
  title = {Towards Distance-Adaptive Wireless Charging},
  booktitle = {Proceedings of the 23rd Annual International Conference on Mobile Systems, Applications and Services},
  series = {ACM MobiSys},
  year = {2025},
  address = {Anaheim, California, USA},
  month = jun,
}

SenSys’24
M3Cam: Extreme Super-resolution via Multi-Modal Optical Flow for Mobile Cameras

Yu Lu , Dian Ding , Hao Pan , and 7 more authors

In Proceedings of the 22nd ACM Conference on Embedded Networked Sensor Systems , Nov 2024

Abs Bib

The demand for ultra-high-resolution imaging in mobile phone photography is continuously increasing. However, the image resolution of mobile devices is typically constrained by the size of the CMOS sensor. Although deep learning-based super-resolution (SR) techniques have the potential to overcome this limitation, existing SR neural network models require large computational resources, making them unsuitable for real-time SR imaging on current mobile devices. Additionally, cloud-based SR systems pose privacy leakage risks. In this paper, we propose M3Cam, an innovative and lightweight SR imaging system for mobile phones. M3Cam can ensure high-quality 16\texttimes SR image (4\texttimes in both height and width) visualization with almost negligible latency. In detail, we utilize an optical image stabilization (OIS) module for lens control and introduce a new modality of data, namely gyroscope readings, to achieve high-precision and compact optical flow estimation modules. Building upon this concept, we design a multi-frame-based SR model utilizing the Swin Transformer. Our proposed system can generate a 16\texttimes SR image from four captured low-resolution images in real-time, with low computational load, low inference latency, and minimal reliance on runtime RAM. Through extensive experiments, we demonstrate that our proposed multi-modal optical flow model significantly enhances pixel alignment accuracy between multiple frames and delivers outstanding 16\texttimes SR imaging results under various shooting scenarios. Code and dataset are available at: https://github.com/liangjindeamo-yuer/M3CAM
@inproceedings{M3Cam, author = {Lu, Yu and Ding, Dian and Pan, Hao and Fu, Yongjian and Zhang, Liyun and Tan, Feitong and Wang, Ran and Chen, Yi-Chao and Xue, Guangtao and Ren, Ju}, title = {M3Cam: Extreme Super-resolution via Multi-Modal Optical Flow for Mobile Cameras}, year = {2024}, isbn = {9798400706974}, publisher = {Association for Computing Machinery}, address = {New York, NY, USA}, url = {https://doi.org/10.1145/3666025.3699371}, doi = {10.1145/3666025.3699371}, booktitle = {Proceedings of the 22nd ACM Conference on Embedded Networked Sensor Systems}, pages = {744–756}, numpages = {13}, keywords = {super-resolution system, optical flow, mobile camera}, location = {Hangzhou, China}, series = {SenSys'24}, month = nov, }
MobiSys’24
Adaptive Metasurface-Based Acoustic Imaging using Joint Optimization

Yongjian Fu, Yongzhao Zhang , Yu Lu , and 7 more authors

In Proceedings of the 22nd Annual International Conference on Mobile Systems, Applications and Services , Jun 2024

Abs Bib

Acoustic imaging is attractive due to its ability to work under occlusion, different lighting conditions, and privacy-sensitive environments. Existing acoustic imaging methods require large transceiver arrays or device movement, which makes it challenging to use in many scenarios. In this paper, we develop a novel acoustic imaging system for low-cost devices with few speakers and microphones without any device movement. To achieve this goal, we leverage a 3D-printed passive acoustic metasurface to significantly enhance the diversity of the measurement data, thereby improving the imaging quality. Specifically, we jointly design the transmission signal, transceivers’ beamforming weights, metasurface, and imaging algorithm to minimize the imaging reconstruction error in an end-to-end manner. We further develop a scheme to dynamically adapt the imaging resolution based on the distance to the target. We implement a system prototype. Using extensive experiments, we show that our system yields high-quality images across a wide range of scenarios.
@inproceedings{MAJIC, author = {Fu, Yongjian and Zhang, Yongzhao and Lu, Yu and Qiu, Lili and Chen, Yi-Chao and Wang, Yezhou and Wang, Mei and Li, Yijie and Ren, Ju and Zhang, Yaoxue}, title = {Adaptive Metasurface-Based Acoustic Imaging using Joint Optimization}, year = {2024}, isbn = {9798400705816}, publisher = {Association for Computing Machinery}, address = {New York, NY, USA}, url = {https://doi.org/10.1145/3643832.3661863}, doi = {10.1145/3643832.3661863}, booktitle = {Proceedings of the 22nd Annual International Conference on Mobile Systems, Applications and Services}, pages = {492–504}, numpages = {13}, keywords = {acoustic imaging, compressive sensing, joint optimization}, location = {Minato-ku, Tokyo, Japan}, series = {MOBISYS '24}, month = jun, }
MobiCom’24
Pushing Wireless Charging from Station to Travel

Lili Chen , Bozhong Yu , Yongjian Fu, and 4 more authors

In Proceedings of the 30th Annual International Conference on Mobile Computing and Networking , Nov 2024

Abs Bib

Wireless charging has achieved promising progress in recent years. However, the severe bottlenecks are the small charging range and poor flexibility. This paper presents ChargeX to enable smart and long-range wireless charging for small mobile devices. ChargeX incorporates emerging smart metasurface into the magnetic resonance coupling-based wireless charging to extend the charging range and accommodates the mobility of charging device. Unlike previous endeavors in metasurface-assisted wireless charging that focused on simulation, ChargeX makes efforts across software and hardware to meet three crucial requirements for a practical wireless charging system: (i) realize high-freedom and accurate metasurface control under the premise of low loss; (ii) obtain real-time feedback from the receiver and make effective manipulation for transmitted magnetic flux; and (iii) generate a proper AC signal source at the desired frequency band. We developed a prototype of ChargeX, and evaluated its performance through controlled experiments and real-world phone charging. Extensive experiments demonstrate the great potential of ChargeX for long-range and flexible wireless charging with a compact receiver design.
@inproceedings{ChargeY, author = {Chen, Lili and Yu, Bozhong and Fu, Yongjian and Ren, Ju and Pan, Hao and Gummeson, Jeremy and Zhang, Yaoxue}, title = {Pushing Wireless Charging from Station to Travel}, year = {2024}, isbn = {9798400704895}, publisher = {Association for Computing Machinery}, address = {New York, NY, USA}, url = {https://doi.org/10.1145/3636534.3649346}, doi = {10.1145/3636534.3649346}, booktitle = {Proceedings of the 30th Annual International Conference on Mobile Computing and Networking}, location = {Washington D.C., DC, USA}, series = {ACM MobiCom '24}, month = nov, }
IMWUT’24
Pushing the Limits of Acoustic Spatial Perception via Incident Angle Encoding

Yongjian Fu, Yongzhao Zhang , Hao Pan , and 7 more authors

Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., May 2024

Abs Bib

With the growing popularity of smart speakers, numerous novel acoustic sensing applications have been proposed for low-frequency human speech and high-frequency inaudible sounds. Spatial information plays a crucial role in these acoustic applications, enabling various location-based services. However, typically commercial microphone arrays face limitations in spatial perception of inaudible sounds due to their sparse array geometries optimized for low-frequency speech. In this paper, we introduce MetaAng, a system designed to augment microphone arrays by enabling wideband spatial perception across both speech signals and inaudible sounds by leveraging the spatial encoding capabilities of acoustic metasurfaces. Our design is grounded in the fact that, while sensitive to high-frequency signals, acoustic metasurfaces are almost non-responsive to low-frequency speech due to significant wavelength discrepancy. This observation allows us to integrate acoustic metasurfaces with sparse array geometry, simultaneously enhancing the spatial perception of high-frequency and low-frequency acoustic signals. To achieve this, we first utilize acoustic metasurfaces and a configuration optimization algorithm to encode the unique features for each incident angle. Then, we propose an unrolling soft thresholding network that employs neural-enhanced priors and compressive sensing for high-accuracy, high-resolution multi-source angle estimation. We implement a prototype, and experimental results demonstrate that MetaAng maintains robustness across various scenarios, facilitating multiple applications, including localization and tracking.
@article{MetaAng_Fu, author = {Fu, Yongjian and Zhang, Yongzhao and Pan, Hao and Lu, Yu and Li, Xinyi and Chen, Lili and Ren, Ju and Li, Xiong and Zhang, Xiaosong and Zhang, Yaoxue}, title = {Pushing the Limits of Acoustic Spatial Perception via Incident Angle Encoding}, year = {2024}, issue_date = {June 2024}, publisher = {Association for Computing Machinery}, address = {New York, NY, USA}, volume = {8}, number = {2}, url = {https://doi.org/10.1145/3659583}, doi = {10.1145/3659583}, journal = {Proc. ACM Interact. Mob. Wearable Ubiquitous Technol.}, month = may, }

SenSys’22

SVoice: Enabling Voice Communication in Silence via Acoustic Sensing on Commodity Devices

Yongjian Fu, Shuning Wang , Linghui Zhong , and 3 more authors

In Proceedings of the 20th ACM Conference on Embedded Networked Sensor Systems , Nov 2022

Bib

@inproceedings{SenSys22_SVoice,
  author = {Fu, Yongjian and Wang, Shuning and Zhong, Linghui and Chen, Lili and Ren, Ju and Zhang, Yaoxue},
  title = {SVoice: Enabling Voice Communication in Silence via Acoustic Sensing on Commodity Devices},
  year = {2022},
  booktitle = {Proceedings of the 20th ACM Conference on Embedded Networked Sensor Systems},
  series = {SenSys'22},
  pages = {622--636},
  month = nov,
  numpages = {15},
  keywords = {acoustic sensing, {cGAN}, silent speech, transformer},
  location = {Boston, Massachusetts},
  publisher = {Association for Computing Machinery},
  url = {https://doi.org/10.1145/3560905.3568530},
  doi = {10.1145/3560905.3568530},
}