67 citations · 299 across the 33 of their papers we have counts for
11 papers · 1 filter
Microsaccade-Inspired Probing: Positional Encoding Perturbations Reveal LLM Misbehaviours
Rui Melo, Rui Abreu, Corina S. Pasareanu
We draw inspiration from microsaccades, tiny involuntary eye movements that reveal hidden dynamics of human perception, to propose an analogous probing method for large language mo…
Toward Certified Robustness Against Real-World Distribution Shifts
Haoze Wu, Teruhiro Tagomori, Alexander Robey +6
We consider the problem of certifying the robustness of deep neural networks against real-world distribution shifts. To do so, we bridge the gap between hand-crafted specifications…
On the Perils of Cascading Robust Classifiers
Ravi Mangal, Zifan Wang, Chi Zhang +3
Ensembling certifiably robust neural networks is a promising approach for improving the \emph{certified robust accuracy} of neural models. Black-box ensembles that assume only quer…
Discrete-Event Controller Synthesis for Autonomous Systems with Deep-Learning Perception Components
Radu Calinescu, Calum Imrie, Ravi Mangal +4
We present DeepDECS, a new method for the synthesis of correct-by-construction discrete-event controllers for autonomous systems that use deep neural network (DNN) classifiers for…
QuantifyML: How Good is my Machine Learning Model?
Muhammad Usman, Divya Gopinath, Corina S. Păsăreanu
The efficacy of machine learning models is typically determined by computing their accuracy on test data sets. However, this may often be misleading, since the test data may not be…
Self-Correcting Neural Networks For Safe Classification
Klas Leino, Aymeric Fromherz, Ravi Mangal +3
Classifiers learnt from data are increasingly being used as components in systems where safety is a critical concern. In this work, we present a formal notion of safety for classif…
DeepCert: Verification of Contextually Relevant Robustness for Neural Network Image Classifiers
Colin Paterson, Haoze Wu, John Grese +3
We introduce DeepCert, a tool-supported method for verifying the robustness of deep neural network (DNN) image classifiers to contextually relevant perturbations such as blur, haze…
NNrepair: Constraint-based Repair of Neural Network Classifiers
Muhammad Usman, Divya Gopinath, Youcheng Sun +2
We present NNrepair, a constraint-based technique for repairing neural network classifiers. The technique aims to fix the logic of the network at an intermediate layer or at the la…
NEUROSPF: A tool for the Symbolic Analysis of Neural Networks
Muhammad Usman, Yannic Noller, Corina Pasareanu +2
This paper presents NEUROSPF, a tool for the symbolic analysis of neural networks. Given a trained neural network model, the tool extracts the architecture and model parameters and…
Fast Geometric Projections for Local Robustness Certification
Aymeric Fromherz, Klas Leino, Matt Fredrikson +2
Local robustness ensures that a model classifies all inputs within an -ball consistently, which precludes various forms of adversarial inputs. In this paper, we present a f…
Property Inference for Deep Neural Networks
Divya Gopinath, Hayes Converse, Corina S. Pasareanu +1
We present techniques for automatically inferring formal properties of feed-forward neural networks. We observe that a significant part (if not all) of the logic of feed forward ne…