Xinguo (Shane) Feng

s.feng@uq.edu.au

About Me

I am an Associate Lecturer at the School of EECS, University of Queensland (UQ). I obtained my PhD from UQ in 2026, advised by Associate Professor Guangdong Bai from CityUHK (previously UQ), Dr Sharif Abuadbba from CSIRO's Data61, and Dr Guowei Yang from UQ.

My research primarily focuses on the privacy and security of Machine Learning and AI systems, exploring potential vulnerabilities and developing robust defenses to protect these systems.

News

[Mar. 2026] Our paper on adaptive model usage control is accepted by IEEE EuroS&P'26.
[Nov. 2025] Our paper on defense on gradient inversion defense is accepted by ACM AsiaCCS'26.
[Jan. 2025] Our paper on neural network model modulation is accepted by ACM WWW'25.
[Jan. 2025] I am invited to serve as a reviewer for Transactions on Dependable and Secure Computing (TDSC).
[Dec. 2024] I am invited to serve as a reviewer for Transactions on Services Computing (TSC).
[Dec. 2024] I am invited to serve as a Program Committee member for ACM WWW'25 Short Paper Track.
[Oct. 2024] I am invited to serve as a Program Committee member for PAKDD'25.
[Oct. 2024] I am invited to serve as a reviewer for Transactions on Information Forensics & Security (TIFS).
[Aug. 2024] Our paper on Gradient Inversion in language model is accepted by ACM CCS’24
[Mar. 2024] Our paper on neural network usage control is accepted by IEEE S&P'24.
[Aug. 2023] Our paper on formalizing language model perturbation is accepted by ICFEM'23.
[Apr. 2023] Our paper on detecting contradictions of IoT documentations is accepted by ISA Transactions.
[Oct. 2022] Our paper on detecting contradictions of CoAP IoT protocol documentation is accepted by NSS'22.

Publications

Re-Key-Free, Risky-Free: Adaptable Model Usage Control

Zihan Wang, Zhongkui Ma, Xinguo Feng, Chuan Yan, Dongge Liu, Ruoxi Sun, Derui Wang, Minhui Xue, Guangdong Bai

IEEE EuroS&P'26

Mitigating Gradient Inversion Risks in Language Models via Token Obfuscation

Xinguo Feng, Zhongkui Ma, Zihan Wang, Alsharif Abuadbba, Guangdong Bai

ACM AsiaCCS'26

AI Model Modulation with Logits Redistribution

Zihan Wang, Zhongkui Ma, Xinguo Feng, Zhiyang Mei, Zhiyong Ma, Derui Wang, Minhui Xue, Guangdong Bai

ACM WWW'25

Uncovering Gradient Inversion Risks in Practical Language Model Training

Xinguo Feng, Zhongkui Ma, Zihan Wang, Eu Joe Chegne, Mengyao Ma, Alsharif Abuadbba, Guangdong Bai

ACM CCS'24

Corelocker: Neuron-level usage control

Zihan Wang, Zhongkui Ma, Xinguo Feng, Ruoxi Sun, Hu Wang, Minhui Xue, Guangdong Bai

IEEE S&P'24

Formalizing Robustness Against Character-Level Perturbations for Neural Network Language Models

Zhongkui Ma, Xinguo Feng, Zihan Wang, Shuofeng Liu, Mengyao Ma, Hao Guan, Mark Huasong Meng

ICFEM'23

Detecting contradictions from IoT protocol specification documents based on neural generated knowledge graph

Xinguo Feng, Yanjun Zhang, Mark Huasong Meng, Yansong Li, Chegne Eu Joe, Zhe Wang, Guangdong Bai

ISA Transactions

Detecting contradictions from coap rfc based on knowledge graph

Xinguo Feng, Yanjun Zhang, Mark Huasong Meng, Sin G Teo

NSS'22