I am a Ph.D. candidate in the School of Computer Science at Northwestern Polytechnical University (NWPU), supervised by Prof. Bin Guo (NWPU) and Academician Yunhao Liu (Tsinghua University, THU), starting from Fall 2021. Before that, I received my B.Eng. in Computer Science and Technology from NWPU (2021) as an Outstanding Graduate.
My research spans Mobile Computing and Resource-Constrained AI (context-aware deep learning on resource-constrained devices, memory-efficient training, cloud–edge collaborative inference and adaptation) and Multimodal Large Language Models (streaming video understanding, hallucination reduction, and anomaly detection).
Contact & profiles: Mail · Google Scholar · GitHub · ORCID
🔥 News
- Apr 2026: 🎉 Our paper Response-G1 was accepted to ACL 2026.
- Mar 2026: 🎉 Our paper TaskIT was accepted to CVPR 2026.
- May 2025: 🎉 Our paper AdaShift was accepted to IEEE TMC.
- Mar 2025: 🎉 Our paper SURGEON was accepted to CVPR 2025 (Highlight, top 3%, 387/13008).
- Mar 2025: 🎉 Our paper CrowdHMTware was accepted to IEEE TMC.
- Jan 2025: 🎉 Our survey Embodied Navigation was accepted to Science China Information Sciences.
- Nov 2024: 🏆 Our paper Adashadow won Best Paper Honorable Mention (7/313) at SenSys 2024.
- Sep 2024: 🎉 Our paper Hawk was accepted to NeurIPS 2024.
- Sep 2024: 🎉 Our paper AdaKnife was accepted to IEEE TMC.
- Aug 2022: 🏆 Our solution for cloud–edge collaborative model adaptation won Huawei Spark Award 2022.
- Apr 2021: 🎉 Our paper AdaSpring was accepted to Ubicomp 2021.
📝 Publications
Selected Research Papers
*: Equal Contribution, †: Corresponding Author.
Response-G1: Explicit Scene Graph Modeling for Proactive Streaming Video Understanding
Ke Ma*, Jiaqi Tang*, Bin Guo†, Xueting Han, Ruonan Xu, Qingfeng He, Ziheng Wang, Xu Wang, Qifeng Chen, Zhiwen Yu, Yunhao Liu†
ACL 2026 (CCF-A)
TaskIT: Memory-Efffcient Fine-Tuning of Multi-LoRA LLMs via Cross-Task Importance Transfer
Cheng Fang, Zimu Zhou†, Ke Ma, Bin Guo
CVPR 2026 (CCF-A)
SURGEON: Memory-Adaptive Fully Test-Time Adaptation via Dynamic Activation Sparsity
Ke Ma, Jiaqi Tang, Bin Guo†, Fan Dang, Sicong Liu, Zhui Zhu, Lei Wu, Cheng Fang, Ying-Cong Chen, Zhiwen Yu, Yunhao Liu†
CVPR 2025 (CCF-A; Highlight, top 3%, 387/13008)
Media: 极市平台
AdaShift: Anti-Collapse and Real-Time Deep Model Evolution for Mobile Vision Applications
Ke Ma, Bin Guo†, Sicong Liu, Cheng Fang, Siqi Luo, Zimu Zheng, Zhiwen Yu
IEEE TMC 2025 (CCF-A, SCI Q1)
AdaShadow: Responsive Test-time Model Adaptation in Non-stationary Mobile Environments
Cheng Fang, Sicong Liu†, Zimu Zhou, Bin Guo†, Jiaqi Tang, Ke Ma, Zhiwen Yu
SenSys 2024 (CCF-B; Best Paper Honorable Mention)
Hawk: Learning to Understand Open-World Video Anomalies
Jiaqi Tang*, Hao Lu*, Ruizheng Wu, Xiaogang Xu, Ke Ma, Cheng Fang, Bin Guo, Jiangbo Lu, Qifeng Chen, Ying-Cong Chen†
NeurIPS 2024 (CCF-A)
Media: 自动驾驶之心
AdaSpring: Context-adaptive and Runtime-evolutionary Deep Model Compression for Mobile Applications
Sicong Liu, Bin Guo, Ke Ma, Zhiwen Yu, Junzhao Du
Ubicomp 2021 (CCF-A)
💻 Internships and Professional Experience
- 2024.11 – 2025.04, AI Engineer (Intern), 2012 Labs, Future Device Department, Huawei, Beijing, China. Mentor: Dr. Zhenzhong Kou.
- Ticketing Engine Optimization: Improved system responsiveness for high-concurrency ticketing apps (e.g., 12306, Damai) on HarmonyOS via DVFS-based performance tuning.
- Emotion-aware Live Wallpaper: Built a real-time Chinese text sentiment model and a quantized HarmonyOS deployment for mood-based wallpapers.
- 2022.09 – 2023.09, Student Researcher (Part-time), Cloud BU, Huawei, Shenzhen, China. Mentor: Dr. Zimu Zheng.
- Cloud–edge Collaborative Model Adaptation: Researched and implemented adaptive algorithms for cloud–edge frameworks under continuous visual domain shift on edge devices.
- 2022.01 – 2023.01, Student Organizer, CrowdHMT, NWPU, Xi’an, China. Mentor: Prof. Bin Guo, Prof. Sicong Liu.
- Crowd Intelligence Computing Platform (Web & Toolkit): Led an integrated platform CrowdHMT. Contributed to Crowdhmtware (IEEE TMC) plus software copyrights and patents.
- Book Project: Co-authored Chapter 8 of Introduction to AIoT, Chapters 3 & 4 of Human–Machine–Thing Crowd Intelligence Computing.
✨ Other Experience
- 2018.09 – 2020.09, Vice President & Minister of Arts, Student Union, NWPU.
- Led arts programming and major events (e.g., Freshmen Welcome Gala, “12.9” patriotic choir). Recognized as Outstanding Student Union Member and Outstanding League Member.
🎖 Honors and Awards
- 2024: Best Paper Honorable Mention, SenSys 2024.
- 2022: Spark Award for Solving World-Class Challenges, Huawei.
- 2021: Outstanding Graduate & Outstanding Thesis Award, NWPU.
- 2020: Honorable Mention (top 20%), Mathematical Contest in Modeling.
- 2020: National Second Prize, National Service Outsourcing Innovation and Entrepreneurship Competition.
🤝 Service and Skills
- Reviewer: Ubicomp (2026-), MM (2026-), ACL (2026-), NeurIPS (2025-), CVPR (2025-), AAAI (2024-).
- Membership: IEEE Student Member, CVF Member, CCF Student Member.
- Programming: Python (PyTorch, TensorFlow, MindSpore), C/C++, Matlab, LATEX.
- Languages: Chinese (Native), English (CET-6 560, CET-4 581).