Ilia Shumailov
- Dynamic / ME-JEPA v2.0.0-rc1: Audited World-Model Runtime and Verified Training-Corpus Artifact
Recursive training on model-generated data causes generative models to lose information about rare events and eventually collapse into inaccurate outputs.
2023/05/27 by Ilia Shumailov, Zakhar Shumaylov, Shumailov, Ilia +9 · 59 voices · 59 citations
Computer Science · #Topic Modeling #Natural Language Processing Techniques #Computational Physics and Python Applications
- Defeating Prompt Injections by Design
2025/03/24 by Edoardo Debenedetti, Debenedetti, Edoardo, Ilia Shumailov +17 · 23 voices · 58 citations
#cs.CR #cs.AI
- Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
2025/07/07 by Gheorghe Comanici, Eric Bieber, Comanici, Gheorghe +6844 · 8 voices · 1345 citations
#cs.CL #cs.AI
- Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
2024/03/08 by Gemini Robotics Team, Petko Georgiev, Gemini Team +2277 · 4 voices · 529 citations
Computer Science · #Semantic Web and Ontologies
- AI models collapse when trained on recursively generated data
2024/07/24 by Ilia Shumailov, Zakhar Shumaylov, Yiren Zhao +3 · 6 voices · 157 citations
Computer Science · #Topic Modeling #Generative Adversarial Networks and Image Synthesis #Domain Adaptation and Few-Shot Learning
- Hearing your touch: A new acoustic side channel on smartphones
2019/03/26 by Ilia Shumailov, Laurent Simon, Shumailov, Ilia +5 · 3 voices · 1 citation
#cs.CR #cs.AI
- Scalable watermarking for identifying large language model outputs
2024/10/23 by Sumanth Dathathri, Abigail See, Sumedh Ghaisas +21 · 1 voice · 80 citations
Computer Science · #Advanced Malware Detection Techniques #Topic Modeling #Internet Traffic Analysis and Secure E-voting
- Bad Characters: Imperceptible NLP Attacks
2021/06/18 by Nicholas Boucher, Ilia Shumailov, Boucher, Nicholas +5 · 1 voice · 6 citations
Computer Science · #Adversarial Robustness in Machine Learning
- The Attacker Moves Second: Stronger Adaptive Attacks Bypass Defenses Against Llm Jailbreaks and Prompt Injections
2025/10/10 by Milad Nasr, Nicholas Carlini, Nasr, Milad +27 · 4 voices · 17 citations
Computer Science · #Adversarial Robustness in Machine Learning #Network Security and Intrusion Detection #Security and Verification in Computing #cs.CR #cs.LG
- Machine Unlearning Doesn't Do What You Think: Lessons for Generative AI Policy and Research
2024/12/09 by A. Feder Cooper, Cooper, A. Feder, Christopher A. Choquette-Choo +75 · 5 voices · 8 citations
Computer Science · Social Sciences · #Ethics and Social Impacts of AI #Explainable Artificial Intelligence (XAI) #cs.AI #cs.CY #cs.LG
- LLM Censorship: A Machine Learning Challenge or a Computer Security Problem?
2023/07/20 by David Glukhov, D. O. Glukhov, Glukhov, David +8 · 1 voice · 6 citations
Computer Science · #Adversarial Robustness in Machine Learning #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #Cryptography and Security (cs.CR) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Topic Modeling #cs.AI #cs.CL #cs.CR #cs.LG
- When the Curious Abandon Honesty: Federated Learning Is Not Private
2021/12/06 by Franziska Boenisch, Boenisch, Franziska, Adam Dziedzic +9 · 13 citations
Computer Science · Medicine · #Privacy-Preserving Technologies in Data #Adversarial Robustness in Machine Learning #Artificial Intelligence in Healthcare and Education
- Inexact Unlearning Needs More Careful Evaluations to Avoid a False Sense of Privacy
2024/03/02 by Jamie Hayes, Ilia Shumailov, Hayes, Jamie +7 · 11 citations
Medicine · Social Sciences · #Cryptography and Security (cs.CR) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Patient Dignity and Privacy #Privacy, Security, and Data Protection
- Measuring memorization in language models via probabilistic extraction
2024/10/25 by Jamie Hayes, Marika Swanberg, Hayes, Jamie +11 · 13 citations
Computer Science · #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning and Algorithms #Natural Language Processing Techniques #Topic Modeling
- Fairness Feedback Loops: Training on Synthetic Data Amplifies Bias
2024/03/12 by Sierra Wyllie, Wyllie, Sierra, Ilia Shumailov +3 · 9 citations
Business, Management and Accounting · Decision Sciences · Social Sciences · #Big Data and Business Intelligence #FOS: Computer and information sciences #Forecasting Techniques and Applications #Machine Learning (cs.LG) #Qualitative Comparative Analysis Research
- Wide Attention Is The Way Forward For Transformers?
2022/10/02 by Jason Ross Brown, Brown, Jason Ross, Yiren Zhao +6 · 1 voice · 1 citation
Computer Science · #Machine Learning and Data Classification #Natural Language Processing Techniques #Topic Modeling #cs.LG
- UnUnlearning: Unlearning is not sufficient for content regulation in advanced generative AI
2024/06/27 by Ilia Shumailov, Shumailov, Ilia, Jamie Hayes +15 · 6 citations
Computer Science · Social Sciences · #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #Cryptography and Security (cs.CR) #Ethics and Social Impacts of AI #FOS: Computer and information sciences #Machine Learning (cs.LG) #Reinforcement Learning in Robotics #Topic Modeling
- Beyond Slow Signs in High-fidelity Model Extraction
2024/06/14 by Hanna Foerster, Robert F. Mullins, Foerster, Hanna +5 · 4 citations
Physics and Astronomy · #Model Reduction and Neural Networks
- Gradients Look Alike: Sensitivity is Often Overestimated in DP-SGD
2023/07/01 by Anvith Thudi, Thudi, Anvith, Hengrui Jia +7 · 3 citations
Computer Science · #Adversarial Robustness in Machine Learning #Artificial Intelligence (cs.AI) #Cryptography and Security (cs.CR) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Privacy-Preserving Technologies in Data #Stochastic Gradient Optimization Techniques
- Buffer Overflow in Mixture of Experts
2024/02/08 by Jamie Hayes, Ilia Shumailov, Hayes, Jamie +3 · 2 citations
Physics and Astronomy · #Complex Network Analysis Techniques
- Architectural Neural Backdoors from First Principles
2024/02/10 by Harry Langford, Langford, Harry, Ilia Shumailov +7 · 2 citations
Computer Science · #Neural Networks and Applications
- Sitatapatra: Blocking the Transfer of Adversarial Samples
2019/01/23 by Ilia Shumailov, Xitong Gao, Shumailov, Ilia +9 · 1 citation
Computer Science · Engineering · #Adversarial Robustness in Machine Learning #Cryptography and Security (cs.CR) #FOS: Computer and information sciences #Integrated Circuits and Semiconductor Failure Analysis #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Physical Unclonable Functions (PUFs) and Hardware Security
- Hardware and Software Platform Inference
2024/11/07 by Cheng Zhang, Zhang, Cheng, Hanna Foerster +7 · 2 citations
Computer Science · Decision Sciences · #Embedded Systems Design Techniques #FOS: Computer and information sciences #Machine Learning (cs.LG) #Scientific Computing and Data Management #Simulation Techniques and Applications
- Large Language Models Can Verbatim Reproduce Long Malicious Sequences
2025/03/21 by Sharon Lin, Lin, Sharon, Krishnamurthy +9 · 1 voice
Computer Science · #FOS: Computer and information sciences #Machine Learning (cs.LG) #cs.LG
- Tendrils of Crime: Visualizing the Diffusion of Stolen Bitcoins
2019/01/07 by Mansoor Ahmed-Rengers, Ilia Shumailov, Ahmed-Rengers, Mansoor +3 · 1 voice
Computer Science · #Anomaly Detection Techniques and Applications #Blockchain Technology Applications and Security #Computers and Society (cs.CY) #Cryptography and Security (cs.CR) #Data Visualization and Analytics #FOS: Computer and information sciences #Human-Computer Interaction (cs.HC) #cs.CR #cs.CY #cs.HC
- Exploring the limits of strong membership inference attacks on large language models
2025/05/24 by Jamie Hayes, Ilia Shumailov, Hayes, Jamie +29 · 2 voices · 4 citations
Computer Science · #Artificial Intelligence (cs.AI) #Cryptography and Security (cs.CR) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Topic Modeling #cs.AI #cs.CR #cs.LG
- Extracting alignment data in open models
2025/10/21 by Federico Barbero, Barbero, Federico, Xiangming Gu +15 · 1 voice
Computer Science · #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Natural Language Processing Techniques #Software Testing and Debugging Techniques #Topic Modeling
- When Vision Fails: Text Attacks Against ViT and OCR
2023/06/12 by Nicholas Boucher, Jenny Blessing, Boucher, Nicholas +7 · 1 citation
Computer Science · #Advanced Malware Detection Techniques #Adversarial Robustness in Machine Learning #Cryptography and Security (cs.CR) #Digital Media Forensic Detection #FOS: Computer and information sciences #Machine Learning (cs.LG)
- Measuring memorization in RLHF for code completion
2024/06/17 by Aneesh Pappu, Pappu, Aneesh, Billy Porter +5 · 2 citations
Computer Science · Engineering · #Computation and Language (cs.CL) #Embedded Systems Design Techniques #FOS: Computer and information sciences #Machine Learning (cs.LG) #Parallel Computing and Optimization Techniques #Real-time simulation and control systems #Software Engineering (cs.SE)
- Locking Machine Learning Models into Hardware
2024/05/31 by Eleanor Clifford, Clifford, Eleanor, Adhithya Saravanan +13 · 1 citation
Computer Science · #Parallel Computing and Optimization Techniques
- SynthID-Image: Image watermarking at internet scale
2025/10/10 by Sven Gowal, Rudy Bunel, Gowal, Sven +49 · 3 citations
Computer Science · #Advanced Steganography and Watermarking Techniques #Adversarial Robustness in Machine Learning #Artificial Intelligence (cs.AI) #Cryptography and Security (cs.CR) #Digital Media Forensic Detection #FOS: Computer and information sciences
- Interpreting the Repeated Token Phenomenon in Large Language Models
2025/03/11 by Itay Yona, Ilia Shumailov, Yona, Itay +7 · 2 citations
Computer Science · #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #Cryptography and Security (cs.CR) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Natural Language Processing Techniques #Topic Modeling
- Breach By A Thousand Leaks: Unsafe Information Leakage in `Safe' AI Responses
2024/07/02 by D. O. Glukhov, Ziwen Han, Glukhov, David +7 · 1 citation
Computer Science · #Adversarial Robustness in Machine Learning
- Stealing User Prompts from Mixture of Experts
2024/10/30 by Itay Yona, Yona, Itay, Ilia Shumailov +5 · 1 citation
Computer Science · #Data Mining Algorithms and Applications
- ceLLMate: Sandboxing Browser AI Agents
2025/12/14 by L. Meng, Meng, Luoxi, Henry Feng +5 · 2 citations
Computer Science · #Web Application Security Vulnerabilities #Security and Verification in Computing #Spam and Phishing Detection
- Cascading Adversarial Bias from Injection to Distillation in Language Models
2025/05/30 by Harsh Chaudhari, Chaudhari, Harsh, Jamie Hayes +9 · 2 voices · 1 citation
Computer Science · #Adversarial Robustness in Machine Learning #Cryptography and Security (cs.CR) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Natural Language Processing Techniques #Topic Modeling #cs.CR #cs.LG