Ilia Shumailov
13 papers · 2021–2025 · 7 venues · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓+4 more ↓ Show less ↑
🌉 Interdisciplinary Bridge · Silver 🗺️ Taxonomy Completionist (25) 🌍 Conference Polyglot · Silver 🐣 Hot Topic Early Bird 🗃️ Keyword Collector (67)
⚡
Prolific Year · Productive
🔥
Unstoppable · Bronze
❓
The Questioner
(2)
💎
Century Club · Bronze
Top co-authors
Research topics
Artificial Intelligence
(10)
Core AI
(10)
Machine Learning
(9)
Natural Language Processing
(5)
Learning Types · Machine Learning
(5)
Application Areas
(4)
Computer Science
(3)
Deep Learning
(3)
Applications
(3)
Optimization & Theory
(3)
Resources & Methods
(3)
Architectures
(2)
Learning Types · Deep Learning
(2)
Computer Vision
(1)
Generation
(1)
Keywords
adversarial attack
(3)
adversarial machine learning
(2)
llms
(2)
language model
(2)
neural network security
(2)
neural network
(2)
large language model
(2)
code completion
(1)
model architecture
(1)
adversarial learning
(1)
prompt engineering
(1)
reinforcement learning from human feedback
(1)
model alignment
(1)
parameter optimization
(1)
inductive bia
(1)
few-shot learning
(1)
memory efficiency
(1)
black-box attack
(1)
direct preference optimization
(1)
backdoor attack
(1)
Papers
Breach By A Thousand Leaks: Unsafe Information Leakage in 'Safe' AI Responses
ICLR 2025
Hardware and Software Platform Inference
ICML 2025
Interpreting the Repeated Token Phenomenon in Large Language Models
ICML 2025
Measuring memorization in language models via probabilistic extraction
NAACL 2025
Measuring memorization in RLHF for code completion
ICLR 2025
Position: Machine Learning Models Have a Supply Chain Problem
ICML 2025
Beyond Slow Signs in High-fidelity Model Extraction
NeurIPS 2024
Position: Fundamental Limitations of LLM Censorship Necessitate New Approaches
ICML 2024
Architectural Backdoors in Neural Networks
CVPR 2023
Revisiting Automated Prompting: Are We Actually Doing Better?
ACL 2023
Revisiting Block-based Quantisation: What is Important for Sub-8-bit LLM Inference?
EMNLP 2023
Rethinking Image-Scaling Attacks: The Interplay Between Vulnerabilities in Machine Learning Systems
ICML 2022
Markpainting: Adversarial Machine Learning meets Inpainting
ICML 2021