… The reliability and robustness of neural networks can be … , such as coverageguided fuzzing
[14], adversarial-generative fuzzing [15]… [30] presented TensorFuzz, which applied fuzz-based …

… neural networks with deeper layers. We design a series of experiments to evaluate the
CAGFuzz … TensorFuzz [65] is good at automatically discovering errors that result from only a few …

… In this paper, we focus on practical tensor compiler fuzzing and have made the following de…
Second, we propose the first coverage-guided fuzzing approach for testing tensor compilers…

… In this paper, to address these problems, we present a new coverage-guided fuzzer named
… the limitation of previous fuzzing schemes, DeepHunter and TensorFuzz, and derive the …

… To evaluate the effectiveness of CriticalFuzz, we compare it with DeepHunter and
TensorFuzz. Aspects of the comparison include the ability to cover critical neurons and detect …

… Tensorfuzz [13] uses coverage-guided fuzzing idea to detect whether the neural network has
numerical errors, … TensorFuzz does not only detect model classification errors, it shows that …

Recurrent neural networks (RNNs) have been applied to a broad range of applications, … ,
this article aims to develop a coverage-guided testing approach to systematically exploit the …

… At present, the use of fuzzing methods may be an effective exploration direction. We choose
coverage-guided fuzzing as a method to test neural networks, and use neuron coverage as …

… model, ie, a deep neural network trained with a set of training … of its underlying library, when
debugging [44], incurring more … In this work, we propose Muffin, a fuzzing-based approach to …

… TensorFuzz shows that coverage-guided test generation is helpful for debugging and
understanding neural networks. … In contrast, TensorFuzz is based on randomized fuzzing, and its …