Ilia Shumailov

Jamie Hayes

Eleni Triantafillou

Guillermo Ortiz-Jiménez

CoRR, 2024

Measuring memorization in RLHF for code completion.

[BibT_eX]

[DOI]

CoRR, 2024

Beyond Slow Signs in High-fidelity Model Extraction.

[BibT_eX]

[DOI]

CoRR, 2024

Locking Machine Learning Models into Hardware.

[BibT_eX]

[DOI]

CoRR, 2024

Inexact Unlearning Needs More Careful Evaluations to Avoid a False Sense of Privacy.

[BibT_eX]

[DOI]

CoRR, 2024

Architectural Neural Backdoors from First Principles.

[BibT_eX]

[DOI]

CoRR, 2024

Buffer Overflow in Mixture of Experts.

[BibT_eX]

[DOI]

Jamie Hayes

Itay Yona

CoRR, 2024

Gradients Look Alike: Sensitivity is Often Overestimated in DP-SGD.

[BibT_eX]

[DOI]

Proceedings of the 33rd USENIX Security Symposium, 2024

Machine Learning needs Better Randomness Standards: Randomised Smoothing and PRNG-based attacks.

[BibT_eX]

[DOI]

Pranav Dahiya

Ross Anderson

Proceedings of the 33rd USENIX Security Symposium, 2024

ImpNet: Imperceptible and blackbox-undetectable backdoors in compiled neural networks.

[BibT_eX]

[DOI]

Proceedings of the IEEE Conference on Secure and Trustworthy Machine Learning, 2024

Position: Fundamental Limitations of LLM Censorship Necessitate New Approaches.

[BibT_eX]

[DOI]

Proceedings of the Forty-first International Conference on Machine Learning, 2024

Fairness Feedback Loops: Training on Synthetic Data Amplifies Bias.

[BibT_eX]

[DOI]

Sierra Calanda Wyllie

Proceedings of the 2024 ACM Conference on Fairness, Accountability, and Transparency, 2024

2023

Beyond Labeling Oracles: What does it mean to steal ML models?

[BibT_eX]

[DOI]

CoRR, 2023

Human-Producible Adversarial Examples.

[BibT_eX]

[DOI]

CoRR, 2023

SEA: Shareable and Explainable Attribution for Query-based Black-box Attacks.

[BibT_eX]

[DOI]

Yue Gao

Kassem Fawaz

CoRR, 2023

LLM Censorship: A Machine Learning Challenge or a Computer Security Problem?

[BibT_eX]

[DOI]

CoRR, 2023

Machine Learning needs its own Randomness Standard: Randomised Smoothing and PRNG-based attacks.

[BibT_eX]

[DOI]

Pranav Dahiya

CoRR, 2023

When Vision Fails: Text Attacks Against ViT and OCR.

[BibT_eX]

[DOI]

CoRR, 2023

The Curse of Recursion: Training on Generated Data Makes Models Forget.

[BibT_eX]

[DOI]

CoRR, 2023

Is Federated Learning a Practical PET Yet?

[BibT_eX]

[DOI]

Franziska Boenisch

Adam Dziedzic

Roei Schuster

CoRR, 2023

Tubes Among Us: Analog Attack on Automatic Speaker Identification.

[BibT_eX]

[DOI]

Shimaa Ahmed

Yash Wani

Proceedings of the 32nd USENIX Security Symposium, 2023

Boosting Big Brother: Attacking Search Engines with Encodings.

[BibT_eX]

[DOI]

Proceedings of the 26th International Symposium on Research in Attacks, 2023

Reconstructing Individual Data Points in Federated Learning Hardened with Differential Privacy and Secure Aggregation.

[BibT_eX]

[DOI]

Franziska Boenisch

Adam Dziedzic

Roei Schuster

Proceedings of the 8th IEEE European Symposium on Security and Privacy, 2023

When the Curious Abandon Honesty: Federated Learning Is Not Private.

[BibT_eX]

[DOI]

Franziska Boenisch

Adam Dziedzic

Roei Schuster

Proceedings of the 8th IEEE European Symposium on Security and Privacy, 2023

Revisiting Block-based Quantisation: What is Important for Sub-8-bit LLM Inference?

[BibT_eX]

[DOI]

Cheng Zhang

Jianyi Cheng

George A. Constantinides

Yiren Zhao

Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, 2023

Architectural Backdoors in Neural Networks.

[BibT_eX]

[DOI]

Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023

Revisiting Automated Prompting: Are We Actually Doing Better?

[BibT_eX]

[DOI]

Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), 2023

2022

DARTFormer: Finding The Best Type Of Attention.

[BibT_eX]

[DOI]

CoRR, 2022

Wide Attention Is The Way Forward For Transformers.

[BibT_eX]

[DOI]

CoRR, 2022

ImpNet: Imperceptible and blackbox-undetectable backdoors in compiled neural networks.

[BibT_eX]

[DOI]

CoRR, 2022

Augmentation Backdoors.

[BibT_eX]

[DOI]

CoRR, 2022

In Differential Privacy, There is Truth: On Vote Leakage in Ensemble Private Learning.

[BibT_eX]

[DOI]

CoRR, 2022

Efficient Adversarial Training With Data Pruning.

[BibT_eX]

[DOI]

CoRR, 2022

Bounding Membership Inference.

[BibT_eX]

[DOI]

CoRR, 2022

Model Architecture Adaption for Bayesian Neural Networks.

[BibT_eX]

[DOI]

CoRR, 2022

Pipe Overflow: Smashing Voice Authentication for Fun and Profit.

[BibT_eX]

[DOI]

Shimaa Ahmed

Yash Wani

CoRR, 2022

On the Necessity of Auditable Algorithmic Definitions for Machine Unlearning.

[BibT_eX]

[DOI]

Proceedings of the 31st USENIX Security Symposium, 2022

Towards More Robust Keyword Spotting for Voice Assistants.

[BibT_eX]

[DOI]

Proceedings of the 31st USENIX Security Symposium, 2022

Bad Characters: Imperceptible NLP Attacks.

[BibT_eX]

[DOI]

Proceedings of the 43rd IEEE Symposium on Security and Privacy, 2022

Rapid Model Architecture Adaption for Meta-Learning.

[BibT_eX]

[DOI]

Proceedings of the Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems 2022, 2022

In Differential Privacy, There is Truth: on Vote-Histogram Leakage in Ensemble Private Learning.

[BibT_eX]

[DOI]

Proceedings of the Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems 2022, 2022

On the Limitations of Stochastic Pre-processing Defenses.

[BibT_eX]

[DOI]

Proceedings of the Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems 2022, 2022

Rethinking Image-Scaling Attacks: The Interplay Between Vulnerabilities in Machine Learning Systems.

[BibT_eX]

[DOI]

Yue Gao

Kassem Fawaz

Proceedings of the International Conference on Machine Learning, 2022

2021

ExtremeBB: Enabling Large-Scale Research into Extremism, the Manosphere and Their Correlation by Online Forum Data.

[BibT_eX]

[DOI]

CoRR, 2021

Manipulating SGD with Data Ordering Attacks.

[BibT_eX]

[DOI]

Proceedings of the Advances in Neural Information Processing Systems 34: Annual Conference on Neural Information Processing Systems 2021, 2021

Markpainting: Adversarial Machine Learning meets Inpainting.

[BibT_eX]

[DOI]

Proceedings of the 38th International Conference on Machine Learning, 2021

Sponge Examples: Energy-Latency Attacks on Neural Networks.

[BibT_eX]

[DOI]

Proceedings of the IEEE European Symposium on Security and Privacy, 2021

2020

Hey Alexa what did I just type? Decoding smartphone sounds with a voice assistant.

[BibT_eX]

[DOI]

Almos Zarandy

CoRR, 2020

Nudge Attacks on Point-Cloud DNNs.

[BibT_eX]

[DOI]

CoRR, 2020

Not My Deepfake: Towards Plausible Deniability for Machine-Generated Media.

[BibT_eX]

[DOI]

CoRR, 2020

BatNet: Data transmission between smartphones over ultrasound.

[BibT_eX]

[DOI]

Almos Zarandy

CoRR, 2020

Turning Up the Dial: the Evolution of a Cybercrime Market Through Set-up, Stable, and Covid-19 Eras.

[BibT_eX]

[DOI]

Proceedings of the IMC '20: ACM Internet Measurement Conference, 2020

Blackbox Attacks on Reinforcement Learning Agents Using Approximated Temporal Information.

[BibT_eX]

[DOI]

Proceedings of the 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops, 2020

Towards Certifiable Adversarial Sample Detection.

[BibT_eX]

[DOI]

Proceedings of the AISec@CCS 2020: Proceedings of the 13th ACM Workshop on Artificial Intelligence and Security, 2020

2019

Hearing your touch: A new acoustic side channel on smartphones.

[BibT_eX]

[DOI]

CoRR, 2019

Sitatapatra: Blocking the Transfer of Adversarial Samples.

[BibT_eX]

[DOI]

CoRR, 2019

Information Security Meets Adversarial Examples.

[BibT_eX]

[DOI]

Matthias Kirchner

Cecilia Pasquini

Benjamin Maximilian Reinheimer

Proceedings of the IEEE International Workshop on Information Forensics and Security, 2019

Audio CAPTCHA with a Few Cocktails: It's So Noisy I Can't Hear You (Transcript of Discussion).

[BibT_eX]

[DOI]

Fairooz Islam

Benjamin Maximilian Reinheimer

Proceedings of the Security Protocols XXVII, 2019

Audio CAPTCHA with a Few Cocktails: It's so Noisy I Can't Hear You.

[BibT_eX]

[DOI]

Fairooz Islam

Proceedings of the Security Protocols XXVII, 2019

Snitches Get Stitches: On the Difficulty of Whistleblowing (Transcript of Discussion).

[BibT_eX]

[DOI]

Darija Halatova

Proceedings of the Security Protocols XXVII, 2019

Snitches Get Stitches: On the Difficulty of Whistleblowing.

[BibT_eX]

[DOI]

Darija Halatova

Proceedings of the Security Protocols XXVII, 2019

To Compress Or Not To Compress: Understanding The Interactions Between Adversarial Attacks And Neural Network Compression.

[BibT_eX]

[DOI]

Proceedings of the Second Conference on Machine Learning and Systems, SysML 2019, 2019

Mapping the Underground: Supervised Discovery of Cybercrime Supply Chains.

[BibT_eX]

[DOI]

Proceedings of the 2019 APWG Symposium on Electronic Crime Research, 2019

2018

Towards Automatic Discovery of Cybercrime Supply Chains.

[BibT_eX]

[DOI]

CoRR, 2018

The Taboo Trap: Behavioural Detection of Adversarial Samples.

[BibT_eX]

[DOI]

CoRR, 2018

Making Bitcoin Legal.

[BibT_eX]

[DOI]

Ross Anderson

Proceedings of the Security Protocols XXVI, 2018

Tendrils of Crime: Visualizing the Diffusion of Stolen Bitcoins.

[BibT_eX]

[DOI]

Ross Anderson

Proceedings of the 5th International Workshop on Graphical Models for Security, 2018

2017

Computational analysis of valence and arousal in virtual reality gaming using lower arm electromyograms.

[BibT_eX]

[DOI]