Google TensorFlow SparseFillEmptyRows Denial of Service Vulnerability

Google TensorFlow is an end-to-end open source machine learning platform. A security vulnerability exists in Google TensorFlow SparseFillEmptyRows. An attacker can exploit the vulnerability to trigger null pointer dereference in the "tf.raw\u ops.SparseFillEmptyRows" implementation...

Google TensorFlow divide-by-zero error vulnerability (CNVD-2021-36561)

Google TensorFlow is an end-to-end open source machine learning platform. A divide-by-zero error vulnerability exists in the Conv3D implementation in TensorFlow versions prior to 2.5.0. An attacker could exploit the vulnerability to cause the program to crash...

Google TensorFlow heap out-of-bounds read vulnerability (CNVD-2021-37627)

Google TensorFlow is an end-to-end open source machine learning platform. A heap out-of-bounds read vulnerability exists in Google TensorFlow. An attacker can exploit the vulnerability to read data outside the boundaries of the heap allocation buffer in "tf.raw\u ops.QuantizeAndDequantizeV3"...

Google TensorFlow QuantizedReshape Denial of Service Vulnerability

Google TensorFlow is an end-to-end open source machine learning platform. A security vulnerability exists in Google TensorFlow QuantizedReshape. A local attacker could exploit this vulnerability to cause a denial of service...

TensorFlow Denial of Service Vulnerability (CNVD-2021-36542)

Google TensorFlow is an end-to-end open source machine learning platform. A denial of service vulnerability exists in TensorFlow. An attacker can exploit this vulnerability by passing empty images to tf.rawops.DrawBoundingBoxes to cause a CHECK failure denial of service...

Google TensorFlow Conv2DBackpropFilte Denial of Service Vulnerability

Google TensorFlow is an end-to-end open source machine learning platform. A security vulnerability exists in Google TensorFlow Conv2DBackpropFilte. An attacker could exploit the vulnerability to cause a denial of service...

Google TensorFlow StringNGrams Denial of Service Vulnerability (CNVD-2021-37619)

Google TensorFlow is an end-to-end open source machine learning platform. A security vulnerability exists in Google TensorFlow StringNGrams. An attacker can exploit the vulnerability to trigger a dereference to a null pointer in "tf.raw\u ops.stringrams"...

Denial Of Service (DoS)

tensorflow is vulnerable to denial of service. The vulnerability exists due to division by zero error in tf.rawops.MaxPoolGradWithArgmax which is caused by the failure to validate the batch dimension of the tensor is non-zero before dividing it...

Denial Of Service (DoS)

tensorflow is vulnerable to denial of service. An attacker is able to exploit the vulnerability by calling tf.rawops.ImmutableConst with a dtype of tf.resource or tf.variant resulting in a segfault...

Denial Of Service (DoS)

tensorflow is vulnerable to denial of service. The vulnerability exists due to an undefined behavior in tf.rawops.MaxPool3DGradGrad by dereferencing null pointers backing attacker-supplied empty tensors...

Out-of-Bounds Read

tensorflow has out-of-bound read. The vulnerability exists due to failure to validate the indices used to access elements of input/output arrays...

Denial Of Service (DoS)

tensorflow is vulnerable to denial of service. The vulnerability exists due to MatrixTriangularSolve failing to terminate kernel execution if one validation condition fails...

Out-of-Bounds Read

tensorflow has out-of-bound read. The vulnerability exists due to the lack of validation of axis allowing an attacker to be able to read data from outside of the bounds of heap allocated in the buffer in tf.rawops.QuantizeAndDequantizeV3...

Out-of-Bounds Read

tensorflow has out-of-bound read. The vulnerability exists due to a lack of validation in tf.rawops.Dequantize which allows an attacker to trigger a read from outside the allocated bounds of heap...

Denial Of Service (DoS)

tensorflow is vulnerable to denial of service. An attacker may exploit the vulnerability by passing invalid arguments to tf.rawops.Dilation2DBackpropInput causing an out of bounds write to the heap...

Denial Of Service (DoS)

tensorflow is vulnerable to denial of service. An attacker may exploit the vulnerability by inputting a crafted x tensor which may cause a FPE runtime error in tf.rawops.FusedBatchNorm...

Denial Of Service (DoS)

tensorflow is vulnerable to denial of service. An attacker is able to send a malicious input to cause a memory overflow in ParseAttrValue from nested tensors...

Denial Of Service (DoS)

tensorflow is vulnerable to denial of service. An attacker is able to exploit the vulnerability by inputting a crafted model such that the numsplits would be 0, into the split TFLite operator...

Denial Of Service (DoS)

tensorflow is vulnerable to denial of service. A heap underflow occurs because invalid values for axis argument are accepted in tf.rawops.QuantizeAndDequantizeV2, allowing an attacker to read/write to other data on the heap...

Denial Of Service (DoS)

tensorflow is vulnerable to denial of service. An integer division by zero undefined behavior occurs because it performs division without checking not zero in TFLite computation for size of output after TFLite padding, allowing an attacker to crash the application...