Tiantian Wang 王天天

I am an M.S. student at the University of Tokyo, in the Department of Information Physics and Computing. I received my B.Eng. in Artificial Intelligence from Xidian University, where I was part of the Turing Program.

My research interests include human-centered robotics, multimodal sensing, and machine learning. I work with Wi-Fi CSI, wearable IMUs, and vision to understand human activity. Previously, I studied RF signal processing and automatic modulation classification.

Email GitHub

Selected publications

WiDistill framework: random sampling, expert trajectory generation, and trajectory matching

WiDistill: Distilling Large-scale Wi-Fi Datasets with Trajectory Matching

First author

arXiv, 2024

Dataset distillation with trajectory matching for efficient training and cross-user generalization in Wi-Fi sensing.

Paper Figure

Adaptive Wavelet Network architecture with multi-level decomposition and attention

Towards the Automatic Modulation Classification with Adaptive Wavelet Network

Second author

IEEE Transactions on Cognitive Communications and Networking

Automatic modulation classification through multi-level wavelet decomposition and attention.

Figure

AMC-Net architecture with adaptive correction, multi-scale, and feature fusion modules

AMC-NET: An Effective Network for Automatic Modulation Classification

Second author

IEEE International Conference on Acoustics, Speech and Signal Processing

Automatic modulation classification with adaptive correction, multi-scale features, and feature fusion.

Figure

Education

The University of Tokyo

Oct. 2026 to Oct. 2028 (expected)

M.S. in Information Science and Technology

Department of Information Physics and Computing

Xidian University

Sep. 2020 to Jul. 2024

B.Eng. in Artificial Intelligence, Turing Program

Selected among 40 students from 1,450 undergraduates. GPA: 3.7/4.0.

Research experience

2026–2028

Human-centered robotics and human-computer interaction

The University of Tokyo

Multimodal perception, shared autonomy, robotic augmentation, and adaptive human-machine collaboration.

2023–2026

Multimodal human sensing and activity recognition

Wi-Fi CSI, wearable IMUs, computer vision, human identification, 3D pose estimation, and dataset distillation.

2021–2023

RF signal processing and machine learning

Key Laboratory of Intelligent Perception and Image Understanding

Deep learning and time-frequency representations for robust automatic modulation classification.

Awards

Competition background chart showing strain sensitivity versus gravitational-wave frequency for pulsars — Scientific background figure from the competition overview.

G2Net Detecting Continuous Gravitational Waves

Silver medal45th out of 936 teams (top 5%), 2023

The competition focused on detecting weak, long-lasting gravitational-wave signals from rapidly spinning neutron stars in noisy data. This search could help scientists understand the structure of these extreme stars.

Hosted on Kaggle by the European Gravitational Observatory.

Competition

View medal result

Kaggle result: silver medal, team placed 45th out of 936 teams — Kaggle result: silver medal, 45th out of 936 teams.

Notes

I keep a small archive of older posts about learning and everyday life.

Read the old blog