Ilia Shumailov

Publication Activity (10 Years)

Years Active: 2017-2024
Publications (10 Years): 89

Top Topics

Buffer Overflow

Keyword Spotting

Differential Privacy

Top Venues

Security Protocols Workshop

USENIX Security Symposium

Publications

Jamie Hayes, Ilia Shumailov, Eleni Triantafillou, Amr Khalifa, Nicolas Papernot
Inexact Unlearning Needs More Careful Evaluations to Avoid a False Sense of Privacy. CoRR (2024)
Eleanor Clifford, Adhithya Saravanan, Harry Langford, Cheng Zhang, Yiren Zhao, Robert D. Mullins, Ilia Shumailov, Jamie Hayes
Locking Machine Learning Models into Hardware. CoRR (2024)
Sierra Calanda Wyllie, Ilia Shumailov, Nicolas Papernot
Fairness Feedback Loops: Training on Synthetic Data Amplifies Bias. FAccT (2024)
David Glukhov, Ziwen Han, Ilia Shumailov, Vardan Papyan, Nicolas Papernot
A False Sense of Safety: Unsafe Information Leakage in 'Safe' AI Responses. CoRR (2024)
Ilia Shumailov, Jamie Hayes, Eleni Triantafillou, Guillermo Ortiz-Jiménez, Nicolas Papernot, Matthew Jagielski, Itay Yona, Heidi Howard, Eugene Bagdasaryan
UnUnlearning: Unlearning is not sufficient for content regulation in advanced generative AI. CoRR (2024)
Anvith Thudi, Ilia Shumailov, Franziska Boenisch, Nicolas Papernot
From Differential Privacy to Bounds on Membership Inference: Less can be More. Trans. Mach. Learn. Res. 2024 (2024)
Pranav Dahiya, Ilia Shumailov, Ross Anderson
Machine Learning needs Better Randomness Standards: Randomised Smoothing and PRNG-based attacks. USENIX Security Symposium (2024)
Eleanor Clifford, Ilia Shumailov, Yiren Zhao, Ross J. Anderson, Robert D. Mullins
ImpNet: Imperceptible and blackbox-undetectable backdoors in compiled neural networks. SaTML (2024)
Hanna Foerster, Robert D. Mullins, Ilia Shumailov, Jamie Hayes
Beyond Slow Signs in High-fidelity Model Extraction. CoRR (2024)
Aneesh Pappu, Billy Porter, Ilia Shumailov, Jamie Hayes
Measuring memorization in RLHF for code completion. CoRR (2024)
Jamie Hayes, Ilia Shumailov, Itay Yona
Buffer Overflow in Mixture of Experts. CoRR (2024)
Ilia Shumailov, Zakhar Shumaylov, Yiren Zhao, Nicolas Papernot, Ross J. Anderson, Yarin Gal
AI models collapse when trained on recursively generated data. Nat. 631 (8022) (2024)
Sierra Calanda Wyllie, Ilia Shumailov, Nicolas Papernot
Fairness Feedback Loops: Training on Synthetic Data Amplifies Bias. CoRR (2024)
Anvith Thudi, Hengrui Jia, Casey Meehan, Ilia Shumailov, Nicolas Papernot
Gradients Look Alike: Sensitivity is Often Overestimated in DP-SGD. USENIX Security Symposium (2024)
Harry Langford, Ilia Shumailov, Yiren Zhao, Robert D. Mullins, Nicolas Papernot
Architectural Neural Backdoors from First Principles. CoRR (2024)
Franziska Boenisch, Adam Dziedzic, Roei Schuster, Ali Shahin Shamsabadi, Ilia Shumailov, Nicolas Papernot
When the Curious Abandon Honesty: Federated Learning Is Not Private. EuroS&P (2023)
Cheng Zhang, Jianyi Cheng, Ilia Shumailov, George A. Constantinides, Yiren Zhao
Revisiting Block-based Quantisation: What is Important for Sub-8-bit LLM Inference? CoRR (2023)
Yue Gao, Ilia Shumailov, Kassem Fawaz
SEA: Shareable and Explainable Attribution for Query-based Black-box Attacks. CoRR (2023)
Nicholas Boucher, Jenny Blessing, Ilia Shumailov, Ross Anderson, Nicolas Papernot
When Vision Fails: Text Attacks Against ViT and OCR. CoRR (2023)
Nicholas Boucher, Luca Pajola, Ilia Shumailov, Ross J. Anderson, Mauro Conti
Boosting Big Brother: Attacking Search Engines with Encodings. RAID (2023)
Ilia Shumailov, Zakhar Shumaylov, Yiren Zhao, Yarin Gal, Nicolas Papernot, Ross J. Anderson
The Curse of Recursion: Training on Generated Data Makes Models Forget. CoRR (2023)
Mikel Bober-Irizar, Ilia Shumailov, Yiren Zhao, Robert D. Mullins, Nicolas Papernot
Architectural Backdoors in Neural Networks. CVPR (2023)
Anvith Thudi, Hengrui Jia, Casey Meehan, Ilia Shumailov, Nicolas Papernot
Gradients Look Alike: Sensitivity is Often Overestimated in DP-SGD. CoRR (2023)
Shimaa Ahmed, Yash Wani, Ali Shahin Shamsabadi, Mohammad Yaghini, Ilia Shumailov, Nicolas Papernot, Kassem Fawaz
Tubes Among Us: Analog Attack on Automatic Speaker Identification. USENIX Security Symposium (2023)
Pranav Dahiya, Ilia Shumailov, Ross J. Anderson
Machine Learning needs its own Randomness Standard: Randomised Smoothing and PRNG-based attacks. CoRR (2023)
Yulin Zhou, Yiren Zhao, Ilia Shumailov, Robert D. Mullins, Yarin Gal
Revisiting Automated Prompting: Are We Actually Doing Better? ACL (2) (2023)
Avital Shafran, Ilia Shumailov, Murat A. Erdogdu, Nicolas Papernot
Beyond Labeling Oracles: What does it mean to steal ML models? CoRR (2023)
Franziska Boenisch, Adam Dziedzic, Roei Schuster, Ali Shahin Shamsabadi, Ilia Shumailov, Nicolas Papernot
Is Federated Learning a Practical PET Yet? CoRR (2023)
David Glukhov, Ilia Shumailov, Yarin Gal, Nicolas Papernot, Vardan Papyan
LLM Censorship: A Machine Learning Challenge or a Computer Security Problem? CoRR (2023)
Nicholas Boucher, Luca Pajola, Ilia Shumailov, Ross J. Anderson, Mauro Conti
Boosting Big Brother: Attacking Search Engines with Encodings. CoRR (2023)
Cheng Zhang, Jianyi Cheng, Ilia Shumailov, George A. Constantinides, Yiren Zhao
Revisiting Block-based Quantisation: What is Important for Sub-8-bit LLM Inference? EMNLP (2023)
David Khachaturov, Yue Gao, Ilia Shumailov, Robert D. Mullins, Ross J. Anderson, Kassem Fawaz
Human-Producible Adversarial Examples. CoRR (2023)
Franziska Boenisch, Adam Dziedzic, Roei Schuster, Ali Shahin Shamsabadi, Ilia Shumailov, Nicolas Papernot
Reconstructing Individual Data Points in Federated Learning Hardened with Differential Privacy and Secure Aggregation. EuroS&P (2023)
Yulin Zhou, Yiren Zhao, Ilia Shumailov, Robert Mullins, Yarin Gal
Revisiting Automated Prompting: Are We Actually Doing Better? CoRR (2023)
Jiaqi Wang, Roei Schuster, Ilia Shumailov, David Lie, Nicolas Papernot
In Differential Privacy, There is Truth: On Vote Leakage in Ensemble Private Learning. CoRR (2022)
Anvith Thudi, Ilia Shumailov, Franziska Boenisch, Nicolas Papernot
Bounding Membership Inference. CoRR (2022)
Maximilian Kaufmann, Yiren Zhao, Ilia Shumailov, Robert Mullins, Nicolas Papernot
Efficient Adversarial Training With Data Pruning. CoRR (2022)
Shimaa Ahmed, Yash Wani, Ali Shahin Shamsabadi, Mohammad Yaghini, Ilia Shumailov, Nicolas Papernot, Kassem Fawaz
Pipe Overflow: Smashing Voice Authentication for Fun and Profit. CoRR (2022)
Mikel Bober-Irizar, Ilia Shumailov, Yiren Zhao, Robert Mullins, Nicolas Papernot
Architectural Backdoors in Neural Networks. CoRR (2022)
Anvith Thudi, Hengrui Jia, Ilia Shumailov, Nicolas Papernot
On the Necessity of Auditable Algorithmic Definitions for Machine Unlearning. USENIX Security Symposium (2022)
Jason Ross Brown, Yiren Zhao, Ilia Shumailov, Robert D. Mullins
DARTFormer: Finding The Best Type Of Attention. CoRR (2022)
Yue Gao, Ilia Shumailov, Kassem Fawaz, Nicolas Papernot
On the Limitations of Stochastic Pre-processing Defenses. CoRR (2022)
Yiren Zhao, Xitong Gao, Ilia Shumailov, Nicolò Fusi, Robert D. Mullins
Rapid Model Architecture Adaption for Meta-Learning. NeurIPS (2022)
Shimaa Ahmed, Ilia Shumailov, Nicolas Papernot, Kassem Fawaz
Towards More Robust Keyword Spotting for Voice Assistants. USENIX Security Symposium (2022)
Yue Gao, Ilia Shumailov, Kassem Fawaz, Nicolas Papernot
On the Limitations of Stochastic Pre-processing Defenses. NeurIPS (2022)
Tim Clifford, Ilia Shumailov, Yiren Zhao, Ross J. Anderson, Robert Mullins
ImpNet: Imperceptible and blackbox-undetectable backdoors in compiled neural networks. CoRR (2022)
Jason Ross Brown, Yiren Zhao, Ilia Shumailov, Robert D. Mullins
Wide Attention Is The Way Forward For Transformers. CoRR (2022)
Joseph Rance, Yiren Zhao, Ilia Shumailov, Robert Mullins
Augmentation Backdoors. CoRR (2022)
Yue Gao, Ilia Shumailov, Kassem Fawaz
Rethinking Image-Scaling Attacks: The Interplay Between Vulnerabilities in Machine Learning Systems. ICML (2022)
Nicholas Boucher, Ilia Shumailov, Ross Anderson, Nicolas Papernot
Bad Characters: Imperceptible NLP Attacks. IEEE Symposium on Security and Privacy (2022)
Jiaqi Wang, Roei Schuster, Ilia Shumailov, David Lie, Nicolas Papernot
In Differential Privacy, There is Truth: on Vote-Histogram Leakage in Ensemble Private Learning. NeurIPS (2022)
Duo Wang, Yiren Zhao, Ilia Shumailov, Robert D. Mullins
Model Architecture Adaption for Bayesian Neural Networks. CoRR (2022)
David Khachaturov, Ilia Shumailov, Yiren Zhao, Nicolas Papernot, Ross J. Anderson
Markpainting: Adversarial Machine Learning meets Inpainting. ICML (2021)
Ilia Shumailov, Zakhar Shumaylov, Dmitry Kazhdan, Yiren Zhao, Nicolas Papernot, Murat A. Erdogdu, Ross J. Anderson
Manipulating SGD with Data Ordering Attacks. NeurIPS (2021)
David Khachaturov, Ilia Shumailov, Yiren Zhao, Nicolas Papernot, Ross J. Anderson
Markpainting: Adversarial Machine Learning meets Inpainting. CoRR (2021)
Anvith Thudi, Hengrui Jia, Ilia Shumailov, Nicolas Papernot
On the Necessity of Auditable Algorithmic Definitions for Machine Unlearning. CoRR (2021)
Nicholas Boucher, Ilia Shumailov, Ross J. Anderson, Nicolas Papernot
Bad Characters: Imperceptible NLP Attacks. CoRR (2021)
Yiren Zhao, Xitong Gao, Ilia Shumailov, Nicolò Fusi, Robert D. Mullins
Rapid Model Architecture Adaption for Meta-Learning. CoRR (2021)
Ilia Shumailov, Yiren Zhao, Daniel Bates, Nicolas Papernot, Robert D. Mullins, Ross Anderson
Sponge Examples: Energy-Latency Attacks on Neural Networks. EuroS&P (2021)
Anh V. Vu, Lydia Wilson, Yi Ting Chua, Ilia Shumailov, Ross J. Anderson
ExtremeBB: Enabling Large-Scale Research into Extremism, the Manosphere and Their Correlation by Online Forum Data. CoRR (2021)
Franziska Boenisch, Adam Dziedzic, Roei Schuster, Ali Shahin Shamsabadi, Ilia Shumailov, Nicolas Papernot
When the Curious Abandon Honesty: Federated Learning Is Not Private. CoRR (2021)
Ilia Shumailov, Zakhar Shumaylov, Dmitry Kazhdan, Yiren Zhao, Nicolas Papernot, Murat A. Erdogdu, Ross J. Anderson
Manipulating SGD with Data Ordering Attacks. CoRR (2021)
Ilia Shumailov, Yiren Zhao, Daniel Bates, Nicolas Papernot, Robert D. Mullins, Ross J. Anderson
Sponge Examples: Energy-Latency Attacks on Neural Networks. CoRR (2020)
Ilia Shumailov, Yiren Zhao, Robert D. Mullins, Ross J. Anderson
Towards Certifiable Adversarial Sample Detection. AISec@CCS (2020)
Almos Zarandy, Ilia Shumailov, Ross J. Anderson
Hey Alexa what did I just type? Decoding smartphone sounds with a voice assistant. CoRR (2020)
Yiren Zhao, Ilia Shumailov, Han Cui, Xitong Gao, Robert D. Mullins, Ross J. Anderson
Blackbox Attacks on Reinforcement Learning Agents Using Approximated Temporal Information. DSN Workshops (2020)
Ilia Shumailov, Yiren Zhao, Robert D. Mullins, Ross J. Anderson
Towards Certifiable Adversarial Sample Detection. CoRR (2020)
Mansoor Ahmed-Rengers, Ross J. Anderson, Darija Halatova, Ilia Shumailov
Snitches Get Stitches: On The Difficulty of Whistleblowing. CoRR (2020)
Anh V. Vu, Jack Hughes, Ildiko Pete, Ben Collier, Yi Ting Chua, Ilia Shumailov, Alice Hutchings
Turning Up the Dial: the Evolution of a Cybercrime Market Through Set-up, Stable, and Covid-19 Eras. Internet Measurement Conference (2020)
Baiwu Zhang, Jin Peng Zhou, Ilia Shumailov, Nicolas Papernot
Not My Deepfake: Towards Plausible Deniability for Machine-Generated Media. CoRR (2020)
Yiren Zhao, Ilia Shumailov, Robert D. Mullins, Ross J. Anderson
Nudge Attacks on Point-Cloud DNNs. CoRR (2020)
Almos Zarandy, Ilia Shumailov, Ross J. Anderson
BatNet: Data transmission between smartphones over ultrasound. CoRR (2020)
Yiren Zhao, Ilia Shumailov, Robert D. Mullins, Ross Anderson
To Compress Or Not To Compress: Understanding The Interactions Between Adversarial Attacks And Neural Network Compression. MLSys (2019)
Yiren Zhao, Ilia Shumailov, Han Cui, Xitong Gao, Robert D. Mullins, Ross J. Anderson
Blackbox Attacks on Reinforcement Learning Agents Using Approximated Temporal Information. CoRR (2019)
Rasika Bhalerao, Maxwell Aliapoulios, Ilia Shumailov, Sadia Afroz, Damon McCoy
Mapping the Underground: Supervised Discovery of Cybercrime Supply Chains. eCrime (2019)
Ilia Shumailov, Laurent Simon, Jeff Yan, Ross Anderson
Hearing your touch: A new acoustic side channel on smartphones. CoRR (2019)
Benjamin Maximilian Reinheimer, Fairooz Islam, Ilia Shumailov
Audio CAPTCHA with a Few Cocktails: It's so Noisy I Can't Hear You. Security Protocols Workshop (2019)
Ilia Shumailov, Xitong Gao, Yiren Zhao, Robert D. Mullins, Ross Anderson, Cheng-Zhong Xu
Sitatapatra: Blocking the Transfer of Adversarial Samples. CoRR (2019)
Mansoor Ahmed-Rengers, Ross J. Anderson, Darija Halatova, Ilia Shumailov
Snitches Get Stitches: On the Difficulty of Whistleblowing (Transcript of Discussion). Security Protocols Workshop (2019)
Mansoor Ahmed-Rengers, Ilia Shumailov, Ross Anderson
Tendrils of Crime: Visualizing the Diffusion of Stolen Bitcoins. CoRR (2019)
Benjamin Maximilian Reinheimer, Fairooz Islam, Ilia Shumailov
Audio CAPTCHA with a Few Cocktails: It's So Noisy I Can't Hear You (Transcript of Discussion). Security Protocols Workshop (2019)
Matthias Kirchner, Cecilia Pasquini, Ilia Shumailov
Information Security Meets Adversarial Examples. WIFS (2019)
Mansoor Ahmed-Rengers, Ross J. Anderson, Darija Halatova, Ilia Shumailov
Snitches Get Stitches: On the Difficulty of Whistleblowing. Security Protocols Workshop (2019)
Ross Anderson, Ilia Shumailov, Mansoor Ahmed-Rengers
Making Bitcoin Legal. Security Protocols Workshop (2018)
Ilia Shumailov, Yiren Zhao, Robert D. Mullins, Ross Anderson
The Taboo Trap: Behavioural Detection of Adversarial Samples. CoRR (2018)
Yiren Zhao, Ilia Shumailov, Robert D. Mullins, Ross Anderson
To compress or not to compress: Understanding the Interactions between Adversarial Attacks and Neural Network Compression. CoRR (2018)
Rasika Bhalerao, Maxwell Aliapoulios, Ilia Shumailov, Sadia Afroz, Damon McCoy
Towards Automatic Discovery of Cybercrime Supply Chains. CoRR (2018)
Mansoor Ahmed-Rengers, Ilia Shumailov, Ross Anderson
Tendrils of Crime: Visualizing the Diffusion of Stolen Bitcoins. GraMSec@FLoC (2018)
Ilia Shumailov, Hatice Gunes
Computational analysis of valence and arousal in virtual reality gaming using lower arm electromyograms. ACII (2017)