Full List of Publications

← Back to homepage

Full publication list of Dongqi Cai (蔡栋琪), newest first. * = equal contribution, # = corresponding author. See also my Google Scholar profile.

---

2026

• [Nature Communications’26] Auditing Unauthorized Training Data from AI Generated Content Using Information Isotopes

  Tao Qi, Jinhua Yin, Dongqi Cai, Yueqi Xie, Huili Wang, Zhiyang Hu, Peiru Yang, Guoshun Nan, Zhili Zhou, Chuhan Wu, Lingjuan Lyu, Shangguang Wang, Yongfeng Huang, Nicholas D. Lane

• [IEEE TMC’26] FwdLLM+: Accelerating Forward-only FedLLM with Low-rank Perturbations

  Chen Peng, Mengwei Xu, Zhenyan Lu, Wei Liu, Shangguang Wang, Nicholas D. Lane, Qibo Sun, Dongqi Cai#

• [Science China Information Sciences’26] Resource-Efficient Federated Fine-Tuning for Privacy-Preserving NLP Applications

  Dongqi Cai, Shangguang Wang, Yaozong Wu, Hanlin Gu, Yifan Duan, Lixin Fan, Nicholas D. Lane, Mengwei Xu

• [Patterns’26] Tina: A Diffusion Neural Network for Generating Personalized AI Models from Text Prompts

  Zexi Li, Lingzhi Gao, Dongqi Cai, Nicholas D. Lane, Chao Wu

• [ICLR’26] MobiEdit: Resource-efficient Knowledge Editing for Personalized On-device LLMs [MobiUK’25 Best Poster Award]

  Zhenyan Lu, Daliang Xu, Dongqi Cai, Zexi Li, Wei Liu, Fangming Liu, Shangguang Wang, Mengwei Xu

• [AAAI’26] Editing as Unlearning: Are Knowledge Editing Methods Strong Baselines for Large Language Model Unlearning?

  Zexi Li, Xiangzhu Wang, William F. Shen, Meghdad Kurmanji, Xinchi Qiu, Dongqi Cai, Chao Wu, Nicholas D. Lane

• [arXiv’26] Beyond Scaling: Agents Are Heading to the Edge

  Chunlin Tian*, Dongqi Cai*, Wanru Zhao*, Nicholas D. Lane

2025

• [Nature Communications’25] Ubiquitous Memory Augmentation via Mobile Multimodal Embedding System [pdf] [supplementary] [peer review]

  Dongqi Cai, Shangguang Wang, Chen Peng, Zeling Zhang, Zhenyan Lu, Tao Qi, Nicholas D. Lane, Mengwei Xu

• [Acta Electronica Sinica’25 (电子学报)] 面向微控制单元的高效语音隐私保护编码器 (Efficient and Privacy-Preserving Spoken Language Understanding for Resource-Constrained Microcontroller Units)

  Dongqi Cai, Shangguang Wang, Zeling Zhang, Xiao Ma, Mengwei Xu

• [ACM Computing Surveys’25] Resource-efficient Algorithms and Systems of Foundation Models: A Survey [pdf] [arXiv] [paper list]

  Mengwei Xu*, Dongqi Cai*, Wangsong Yin*, Shangguang Wang, Xin Jin, Xuanzhe Liu

• [ACL’25] Demystifying Small Language Models for Edge Deployment [pdf] [arXiv] [leaderboard]

  Zhenyan Lu, Xiang Li, Dongqi Cai, Rongjie Yi, Fangming Liu, Wei Liu, Jian Luan, Xiwen Zhang, Nicholas D. Lane, Mengwei Xu

• [ICLR’25 Oral, top 1.8%] DEPT: Decoupled Embeddings for Pre-training Language Models

  Alex Iacob, Lorenzo Sani, Meghdad Kurmanji, William F. Shen, Xinchi Qiu, Dongqi Cai, Yan Gao, Nicholas D. Lane

• [ICLR’25] ShortcutsBench: A Large-Scale Real-world Benchmark for API-based Agents

  Haiyang Shen, Yue Li, Desong Meng, Dongqi Cai, Sheng Qi, Li Zhang, Mengwei Xu, Yun Ma

• [MLSys’25] Photon: Federated Pre-training of Large Language Models

  Lorenzo Sani, Alex Iacob, Zeyu Cao, Royson Lee, Bill Marino, Yan Gao, Wanru Zhao, Dongqi Cai, Zexi Li, Xinchi Qiu, Nicholas D. Lane

2024

• [IEEE TBD’24] Accelerating Vertical Federated Learning [pdf]

  Dongqi Cai, Tao Fan, Yan Kang, Lixin Fan, Mengwei Xu, Shangguang Wang, Qiang Yang

• [NeurIPS’24] SILENCE: Protecting Privacy in Offloaded Speech Understanding on Resource-constrained Devices [pdf] [slides]

  Dongqi Cai, Shangguang Wang, Zeling Zhang, Felix Xiaozhu Lin, Mengwei Xu

• [USENIX ATC’24] FwdLLM: Efficient Federated Finetuning of Large Language Models with Perturbed Inferences [pdf] [code] [slides]

  Mengwei Xu, Dongqi Cai#, Yaozong Wu, Xiang Li, Shangguang Wang

• [MobiCom’24] Mobile Foundation Model as Firmware [pdf] [code] [slides] [Distinguished Artifact Nomination]

  Jinliang Yuan*, Chen Yang*, Dongqi Cai*, Shihe Wang, Xin Yuan, Zeling Zhang, Xiang Li, Dingge Zhang, Hanzi Mei, Xianqing Jia, Shangguang Wang, Mengwei Xu

• [EdgeFM’24 @ MobiSys] Large Language Models on Mobile Devices: Measurements, Analysis, and Insights

  Xiang Li, Zhenyan Lu, Dongqi Cai, Xiao Ma, Mengwei Xu

• [EuroMLSys’24 @ EuroSys] FedRDMA: Communication-Efficient Cross-Silo Federated LLM via Chunked RDMA Transmission [pdf]

  Zeling Zhang*, Dongqi Cai*, Yiran Zhang, Mengwei Xu, Shangguang Wang, Ao Zhou

2023

• [MobiCom’23] Federated Few-shot Learning for Mobile NLP [pdf] [slides] [code]

  Dongqi Cai, Shangguang Wang, Yaozong Wu, Felix Xiaozhu Lin, Mengwei Xu

• [MobiCom’23] Efficient Federated Learning for Modern NLP [pdf] [slides] [code] [Most Cited Paper of MobiCom’23 (1/91)]

  Dongqi Cai, Yaozong Wu, Shangguang Wang, Felix Xiaozhu Lin, Mengwei Xu

• [MLICOM’23] GPT4D: Automatic Cross-Version Linux Driver Upgrade Toolkit

  Borui Yang, Hongyu Li, Dongqi Cai#

• [ACM TURC’23] FedAdapter: Efficient Federated Learning for Mobile NLP [poster]

  Dongqi Cai, Shangguang Wang, Yaozong Wu, Mengwei Xu

• [EuroMLSys’23 @ EuroSys] Towards Practical Few-shot Federated NLP [pdf] [poster] [slides]

  Dongqi Cai, Yaozong Wu, Haitao Yuan, Shangguang Wang, Felix Xiaozhu Lin, Mengwei Xu

2021

• [DataCom’21] Mitigating App Collusion Using Machine Learning [pdf]

  Xuefei Duan, Hua Lu, Jinliang Yuan, Qiyang Zhang, Dongqi Cai

• [EMDL’21 @ MobiSys] Towards Ubiquitous Learning: A First Measurement of On-device Training Performance [pdf] [slides]

  Dongqi Cai, Qipeng Wang, Yuanqiang Liu, Yunxin Liu, Shangguang Wang, Mengwei Xu

2020

• [IJTP’20] Implementation of an E-payment Security Evaluation System Based on Quantum Blind Computing [pdf]

  Dongqi Cai, Xi Chen, Yuhong Han, Xin Yi, Jinping Jia, Cong Cao, Ling Fan

---

← Back to homepage