103 citations · 374 across the 18 of their papers we have counts for
18 papers
SelfOp: An Optimization Algorithm for Self-Improving Security Agents
Saad Ullah, Yigitcan Kaya, Christopher Kruegel +2
LLM agents are increasingly used for security tasks: vulnerability discovery, exploit reproduction, and patch generation. Improving them at the model level demands expert demonstra…
From Documentation to Zero-day Vulnerabilities: LLM-Driven Fuzzing of JavaScript Engines in PDF Readers
Suyue Guo, Stijn Pletinckx, Tianle Yu +5
Existing fuzzers for PDF readers rely on simple test cases that involve only individual API calls, leading to limited coverage and potentially missing vulnerabilities that require…
MalwarePT: A Binary-Level Foundation Model for Malware Analysis
Saastha Vasan, Yuzhou Nie, Kaie Chen +6
Automated malware analysis increasingly relies on machine learning, yet most existing methods remain task-specific and depend on handcrafted features or narrowly scoped models. Rec…
When AI Meets the Web: Prompt Injection Risks in Third-Party AI Chatbot Plugins
Yigitcan Kaya, Anton Landerer, Stijn Pletinckx +3
Prompt injection attacks pose a critical threat to large language models (LLMs), with prior work focusing on cutting-edge LLM applications like personal copilots. In contrast, simp…
MADCAT: Combating Malware Detection Under Concept Drift with Test-Time Adaptation
Eunjin Roh, Yigitcan Kaya, Christopher Kruegel +2
We present MADCAT, a self-supervised approach designed to address the concept drift problem in malware detection. MADCAT employs an encoder-decoder architecture and works by test-t…
Like Oil and Water: Group Robustness Methods and Poisoning Defenses May Be at Odds
Michael-Andrei Panaitescu-Liess, Yigitcan Kaya, Sicheng Zhu +2
Group robustness has become a major concern in machine learning (ML) as conventional training paradigms were found to produce high error on minority groups. Without explicit group…
PoisonedParrot: Subtle Data Poisoning Attacks to Elicit Copyright-Infringing Content from Large Language Models
Michael-Andrei Panaitescu-Liess, Pankayaraj Pathmanathan, Yigitcan Kaya +5
As the capabilities of large language models (LLMs) continue to expand, their usage has become increasingly prevalent. However, as reflected in numerous ongoing lawsuits regarding…
ML-Based Behavioral Malware Detection Is Far From a Solved Problem
Yigitcan Kaya, Yizheng Chen, Marcus Botacin +5
Malware detection is a ubiquitous application of Machine Learning (ML) in security. In behavioral malware analysis, the detector relies on features extracted from program execution…
DRSM: De-Randomized Smoothing on Malware Classifier Providing Certified Robustness
Shoumik Saha, Wenxiao Wang, Yigitcan Kaya +2
Machine Learning (ML) models have been utilized for malware detection for over two decades. Consequently, this ignited an ongoing arms race between malware authors and antivirus sy…
Qu-ANTI-zation: Exploiting Quantization Artifacts for Achieving Adversarial Outcomes
Sanghyun Hong, Michael-Andrei Panaitescu-Liess, Yiğitcan Kaya +1
Quantization is a popular technique that the parameter representation of a neural network from floating-point numbers into lower-precision ones (, 8-bit integers…
A Panda? No, It's a Sloth: Slowdown Attacks on Adaptive Multi-Exit Neural Network Inference
Sanghyun Hong, Yiğitcan Kaya, Ionuţ-Vlad Modoranu +1
Recent increases in the computational demands of deep neural networks (DNNs), combined with the observation that most input samples require only simple models, have sparked interes…
On the Effectiveness of Regularization Against Membership Inference Attacks
Yigitcan Kaya, Sanghyun Hong, Tudor Dumitras
Deep learning models often raise privacy concerns as they leak information about their training data. This enables an adversary to determine whether a data point was in a model's t…
On the Effectiveness of Mitigating Data Poisoning Attacks with Gradient Shaping
Sanghyun Hong, Varun Chandrasekaran, Yiğitcan Kaya +2
Machine learning algorithms are vulnerable to data poisoning attacks. Prior taxonomies that focus on specific scenarios, e.g., indiscriminate or targeted, have enabled defenses for…
How to 0wn NAS in Your Spare Time
Sanghyun Hong, Michael Davinroy, Yiğitcan Kaya +2
New data processing pipelines and novel network architectures increasingly drive the success of deep learning. In consequence, the industry considers top-performing architectures a…
Terminal Brain Damage: Exposing the Graceless Degradation in Deep Neural Networks Under Hardware Fault Attacks
Sanghyun Hong, Pietro Frigo, Yiğitcan Kaya +2
Deep neural networks (DNNs) have been shown to tolerate "brain damage": cumulative changes to the network's parameters (e.g., pruning, numerical perturbations) typically result in…
Shallow-Deep Networks: Understanding and Mitigating Network Overthinking
Yigitcan Kaya, Sanghyun Hong, Tudor Dumitras
We characterize a prevalent weakness of deep neural networks (DNNs)---overthinking---which occurs when a DNN can reach correct predictions before its final layer. Overthinking is c…
Security Analysis of Deep Neural Networks Operating in the Presence of Cache Side-Channel Attacks
Sanghyun Hong, Michael Davinroy, Yiǧitcan Kaya +5
Recent work has introduced attacks that extract the architecture information of deep neural networks (DNN), as this knowledge enhances an adversary's capability to conduct black-bo…
Technical Report: When Does Machine Learning FAIL? Generalized Transferability for Evasion and Poisoning Attacks
Octavian Suciu, Radu Mărginean, Yiğitcan Kaya +2
Recent results suggest that attacks against supervised machine learning systems are quite effective, while defenses are easily bypassed by new attacks. However, the specifications…