LipsFlower constrains flow matching to stabilize generative image restoration
LipHDiT uses spectral normalization, bounded modulation and finite-radius gain penalties, tested on blind face restoration.
Stories per day · 30 days
2 RSS
Relatedagent-safety1autonomous-driving1bci1bench2drive1certification1cic-ids20171diffusion-transformer1eeg1eegnet1elf1flow-matching1grad-cam1image-restoration1intrusion-detection1lipschitz1malware-detection1mixture-of-experts1model-compression1
16 stories, newest first · one card per story
Stories tagged #adversarial-robustness
Adversarial robustness does not equal zero-day generalization in intrusion detectors
Perturbation-based robustness tests overestimate zero-day detection; models maintain 90.9-99.7% under attacks but drop to 22.8-76.2% on held-out families.
ORBITAL compresses EEG decoders under a robustness constraint
S25-TRADES reached 23.06% mean true-union robustness versus 21.27% for uncompressed Full EEGNet.
CertMoE certifies MoE LLM robustness under dynamic routing
Existing fixed-graph certificates were violated more than 30 percent of the time; routing-aware checks cut that to zero in tests.
Q-MALDetect study measures evasion of a quantum malware detector
The hybrid detector reached 97.2% training accuracy, but the strongest feature-space attack succeeded 78.7% of the time.
SafeDriveVLA: Navigation-Conditioned World Model Dreaming for Conflict-Aware End-to-End Autonomous Driving
SafeDriveVLA uses world-model dreaming to refuse unsafe driving instructions, cutting collisions from 60 to 38.
TwinViT-DeepJSCC: Adversarially Robust Semantic Image Communication
TwinViT-DeepJSCC defends semantic image links, gaining up to 10.8 dB PSNR under adversarial channel attacks.
Certification of Real Images through Calibrated Content Authentication
Deepfake detectors fall from 99.5% to 76% accuracy; calibrated authentication bounds false certification at 1%.
The Fragility of Trigger-Tag Mechanisms for Misuse Detection in Open-Weight LLMs
Untag attack framework renders existing token- and weight-level trigger-tag misuse detection mechanisms in open-weight LLMs entirely ineffective, researchers demonstrate.
No Free Efficiency: Revisiting the Trade-off Between Training Efficiency and Model Vulnerability
Efficient training recipes make vision and language models easier to attack and less robust.
End-to-End Multimodal Data Augmentation and Adversarial Robustness Benchmark with AugLy for Images, Text, Audio, and PyTorch
A tutorial shows AugLy augmentations and small robustness tests for image, text, and audio in PyTorch.
1
Can Pixels Alone Reveal Image Origin? Minimax Limits and Learnable Interfaces for Passive Provenance
Paper proves pixel-only image provenance has a minimax limit and shows public CLIP verifiers fail under edits.
A Unified and Constrained View of Regularization-Based Robust Reinforcement Learning
Paper unifies regularization-based robust RL methods via new performance-gap upper bounds and jointly learned Lagrange multipliers.
Robust Policy Optimization via Adversarial Importance Sampling
Adversarial Importance Sampling estimates worst-case RL returns without extra interactions; authors also release the advrl PyTorch library.
Certifying Adversarial Robustness of Quantum Classifiers under Known-Readout Query Access
Framework certifies adversarial robustness of quantum classifiers using only measurement statistics and finite-shot outcomes, demonstrated on IBM Quantum hardware.
Do Input-Level Defenses Transfer to Observation-Level Attacks on VideoLLMs?
A systematic study shows input-level adversarial defenses provide inconsistent, often near-zero protection against observation-level attacks on video LLMs.