PT-2026-59975

Impact If Conv2D is given empty input and the filter and padding sizes are valid, the output is all-zeros. This causes division-by-zero floating point exceptions that can be used to trigger a denial of service attack. python import tensorflow as tf import numpy as np with tf.device"CPU": also can...

PT-2026-59845

Impact If Conv2D is given empty input and the filter and padding sizes are valid, the output is all-zeros. This causes division-by-zero floating point exceptions that can be used to trigger a denial of service attack. python import tensorflow as tf import numpy as np with tf.device"CPU": also can...

PYSEC-2026-1029 TensorFlow vulnerable to floating point exception in `Conv2D`

Impact If Conv2D is given empty input and the filter and padding sizes are valid, the output is all-zeros. This causes division-by-zero floating point exceptions that can be used to trigger a denial of service attack. python import tensorflow as tf import numpy as np with tf.device"CPU": also can...

PyTorch DoS

The vulnerabilities of the methods torch.nn.Conv2d, torch.nn.functional.hardshrink, torch.Tensor.view, and torch.Tensor.mv in the PyTorch machine learning framework are related to uncontrolled resource consumption. Exploiting these vulnerabilities could allow a malicious actor to cause service...

tf.keras.layers.Conv2D() DoS

The vulnerability of the tf.keras.layers.Conv2D function in TensorFlow’s machine learning system is related to uncontrolled resource consumption when the padding='valid' parameter is used. Exploiting this vulnerability could allow a malicious actor to cause a service failure...

Google TensorFlow 数字错误漏洞

Google TensorFlow is an end-to-end open source machine learning platform. A divide-by-zero error vulnerability exists in the tf.rawops.Conv2D implementation in TensorFlow versions prior to 2.5.0. No details of the vulnerability are provided at this time...