machine-learning-security

Tag

Cards List
#machine-learning-security

Trojaning the Alignment: Stealthy Backdoor Attacks against Graph Foundation Models

arXiv cs.LG · 5d ago Cached

This paper introduces STAG, a stealthy backdoor attack framework targeting graph foundation models on text-attributed graphs, coordinating graph and text triggers to evade detection and achieve effective attacks.

0 favorites 0 likes
#machine-learning-security

LLM Unlearning for Cyber Defense: A Survey on Methods, Challenges, and Emerging Threats

arXiv cs.LG · 2026-07-21 Cached

This survey examines LLM unlearning methods for cyber defense, introducing a three-level framework to distinguish behavioral suppression, representation-level attenuation, and true forgetting, and analyzing gradient-based, influence-based, and localized editing approaches.

0 favorites 0 likes
#machine-learning-security

ADS-C: Antidistillation Sampling for Classification

arXiv cs.LG · 2026-07-20 Cached

This paper introduces ADS-C, an antidistillation defense for classification that provably preserves top-1 accuracy while degrading student model performance by up to 29.7 percentage points, achieving zero utility cost for the teacher.

0 favorites 0 likes
#machine-learning-security

AESOP: Adversarial Execution-path Selection to Overload Deep Learning Pipelines

arXiv cs.LG · 2026-05-13 Cached

This paper introduces AESOP, a framework for adversarial execution-path selection that significantly inflates FLOPs and latency in deep learning inference pipelines, revealing new efficiency-based vulnerabilities.

0 favorites 0 likes
#machine-learning-security

Attacking machine learning with adversarial examples

OpenAI Blog · 2017-02-24 Cached

This article examines adversarial attacks on machine learning models and demonstrates why gradient masking—a defensive technique that attempts to deny attackers access to useful gradients—is fundamentally ineffective. The paper shows that attackers can circumvent gradient masking by training substitute models that mimic the defended model's behavior, making the defense strategy ultimately futile.

0 favorites 0 likes
← Back to home

Submit Feedback