arXiv AI June 24, 2026 By Merkouris Papamichail, Konstantinos Varsos, Giorgos Flouris, Jo\~ao Marques-Silva

Are Safety Guarantees in Neural Networks Safe? How to Compute Trustworthy Robustness Certifications

arXiv:2606. 23858v1 Announce Type: cross Abstract: A primary challenge in AI safety is the existence of adversarial examples -- slightly distorted inputs that cause a neural network (NN) to misclassify.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv AI

Jul 13

Interval Certifications for Multilayered Perceptrons via Lattice Traversal

arXiv:2607. 08773v1 Announce Type: new Abstract: In this work we present a rigorous theoretical framework to a foundational problem of AI safety, namely adversarial robustness.

By Merkouris Papamichail, Konstantinos Varsos, Giorgos Flouris, Jo\~ao Marques-Silva

safety

arXiv AI

Jul 1

Improving Certified Robustness via Adversarial Distillation

arXiv:2606. 31653v1 Announce Type: cross Abstract: Certified training aims to produce models whose predictions can be formally verified against adversarial perturbations, typically by optimising upper bounds on the worst-case loss over an allowed perturbation set.

By Matteo Melis, Jesus Martinez Del Rincon, Vishal Sharma

efficiency benchmarks safety

arXiv Machine Learning

Aug 19

Certified but Private: Scalable Zero-Knowledge Proofs for Neural Network Guarantees

PANDA is a scalable system that uses zero‑knowledge proofs to certify the robustness and fairness of neural networks without revealing their private parameters. Built on the CROWN robustness framework, PANDA introduces a novel algorithm for proving linear relaxation bounds on non‑linear activation layers, producing lightweight proofs. The system can generate proofs for networks with over 2.9 million parameters in just five minutes and verify them in ten seconds, scaling polynomially with network size and enabling verification of models four orders of magnitude larger than prior ZKP‑based approaches.

By Youwei Zhong, Ben Merbaum, Timos Antonopoulos, Ning Luo, Charalampos Papamanthou, Katerina Sotiraki, Ruzica Piskac

benchmarks safety

arXiv AI

Jun 2

CEAR: Certified Ensemble Adversarial Robustness in DNNs

arXiv:2606. 01437v1 Announce Type: cross Abstract: Deep Neural Networks (DNNs) are highly susceptible to adversarial perturbations, leading to extensive research on robustness for safety-critical applications.

By Daniel Sadig, Mohammadreza Maleki, Hamed Karimi, Reza Samavi

benchmarks safety

arXiv Machine Learning

Jul 1

ShardNet: Training Neural Controllers with Hard, Non-Convex Constraints

arXiv:2606. 30935v1 Announce Type: cross Abstract: While neural network control policies are powerful, their deployment on safety critical systems depends on ensuring that they obey strict constraints.

By Long Kiu Chung, Shreyas Kousik

benchmarks safety

arXiv Machine Learning

Aug 26

$(\text{DNN})^2$: Doubly Non-Negative Relaxations for Deep Neural Networks

arXiv:2608.24743v1 Announce Type: new Abstract: Existing linear program (LP) and semidefinite program (SDP) relaxations for rectified linear unit (ReLU) neural network (NN) verification yield overly-...

By Hanna Jiamei Zhang, Alan Papalia, Michael Everett, David M. Rosen

agents safety

One email a morning, machine-written

Related stories

Interval Certifications for Multilayered Perceptrons via Lattice Traversal

Improving Certified Robustness via Adversarial Distillation

Certified but Private: Scalable Zero-Knowledge Proofs for Neural Network Guarantees

CEAR: Certified Ensemble Adversarial Robustness in DNNs

ShardNet: Training Neural Controllers with Hard, Non-Convex Constraints

$(\text{DNN})^2$: Doubly Non-Negative Relaxations for Deep Neural Networks