Incorrect Calculation of Buffer Size

Overview Affected versions of this package are vulnerable to Incorrect Calculation of Buffer Size when tf.rawops.ResizeNearestNeighborGrad is given a large size input. Remediation Upgrade tensorflow-lite to version 2.12.0 or higher. References - GitHub Commit - Vulnerable Code Credit: Neophytos...

Denial of Service (DoS)

Overview Affected versions of this package are vulnerable to Denial of Service DoS when tf.rawops.TensorListConcat is given elementshape=. Details Denial of Service DoS describes a family of attacks, all aimed at making a system inaccessible to its intended and legitimate users. Unlike other...

Denial of Service (DoS)

Overview Affected versions of this package are vulnerable to Denial of Service DoS when BCast::ToShape is given input larger than an int32, even if it is being supposed to handle up to an int64. Details Denial of Service DoS describes a family of attacks, all aimed at making a system inaccessible...

Buffer Overflow

Overview Affected versions of this package are vulnerable to Buffer Overflow. The reference kernel of the CONV3DTRANSPOSE TensorFlow Lite operator wrongly increments the dataptr when adding the bias to the result. Instead of dataptr += numchannels; it should be dataptr += outputnumchannels; as if...

NULL Pointer Dereference

Overview Affected versions of this package are vulnerable to NULL Pointer Dereference because the pywrap code fails to parse the tensor and returns a nullptr if a list of quantized tensors is assigned to an attribute. Remediation Upgrade tensorflow-lite to version 2.12.0 or higher. References -...

Denial of Service (DoS)

Overview Affected versions of this package are vulnerable to Denial of Service DoS when an input encoded is not a valid CompositeTensorVariant tensor. This will trigger a segfault in tf.rawops.CompositeTensorVariantToComponents. Details Denial of Service DoS describes a family of attacks, all aim...

Denial of Service (DoS)

Overview Affected versions of this package are vulnerable to Denial of Service DoS. This is vulnerable when an input token that is not a UTF-8 bytestring will trigger a CHECK fail in tf.rawops.PyFunc. Details Denial of Service DoS describes a family of attacks, all aimed at making a system...

Out-of-bounds Write

Overview Affected versions of this package are vulnerable to Out-of-bounds Write in FractionalMaxAVGPool with illegal poolingratio. Attackers can access heap memory that is not in the user's control, leading to a crash or remote code execution. Remediation Upgrade tensorflow-lite to version 2.12....

Denial of Service (DoS)

Overview Affected versions of this package are vulnerable to Denial of Service DoS when the input sparsematrix is not a matrix with a shape with rank 0. As a result, a CHECK fail will be triggered in tf.rawops.SparseMatrixNNZ. Details Denial of Service DoS describes a family of attacks, all aimed...

Denial of Service (DoS)

Overview Affected versions of this package are vulnerable to Denial of Service DoS. This is due to the inputs densefeatures or examplestatedata not being of rank 2 which will trigger a CHECK fail in SdcaOptimizer. Details Denial of Service DoS describes a family of attacks, all aimed at making a...

CVE-2022-41894 Buffer overflow in `CONV_3D_TRANSPOSE` on TFLite

TensorFlow is an open source platform for machine learning. The reference kernel of the CONV3DTRANSPOSE TensorFlow Lite operator wrongly increments the dataptr when adding the bias to the result. Instead of dataptr += numchannels; it should be dataptr += outputnumchannels; as if the number of inp...

Google TensorFlow 安全漏洞

Google TensorFlow is an end-to-end open source platform for machine learning from Google, Inc. A security vulnerability exists in Google TensorFlow, which stems from a reference kernel for the CONV3DTRANSPOSE TensorFlow Lite operator that incorrectly adds dataptr when adding bias to results...

CVE-2022-41894

CVE-2022-41894 affects TensorFlow/TFLite CONV_3D_TRANSPOSE reference kernel. The bug increments data_ptr by num_channels instead of output_num_channels, enabling an out-of-bounds write to the bias buffer when input channels exceed output channels. Attack requires using the reference kernel resolv...

PT-2022-26125

Name of the Vulnerable Software and Affected Versions TensorFlow versions prior to 2.11 TensorFlow versions 2.10.1 and earlier TensorFlow versions 2.9.3 and earlier TensorFlow versions 2.8.4 and earlier Description The reference kernel of the CONV 3D TRANSPOSE TensorFlow Lite operator wrongly...

GHSA-79H2-Q768-FPXR TensorFlow segfault TFLite converter on per-channel quantized transposed convolutions

Impact When converting transposed convolutions using per-channel weight quantization the converter segfaults and crashes the Python process. python import tensorflow as tf class QuantConv2DTransposedtf.keras.layers.Layer: def buildself, inputshape: self.kernel = self.addweight"kernel", 3, 3,...

GHSA-FFJM-4QWC-7CMF TensorFlow vulnerable to OOB write in `scatter_nd` in TF Lite

Impact The ScatterNd function takes an input argument that determines the indices of of the output tensor. An input index greater than the output tensor or less than zero will either write content at the wrong index or trigger a crash. Patches We have patched the issue in GitHub commit...

GHSA-PXRW-J2FV-HX3H TensorFlow vulnerable to OOB read in `Gather_nd` in TF Lite

Impact The GatherNd function takes arguments that determine the sizes of inputs and outputs. If the inputs given are greater than or equal to the sizes of the outputs, an out-of-bounds memory read is triggered. Patches We have patched the issue in GitHub commit...

CVE-2022-35937 OOB read in `Gather_nd` op in TensorFlow Lite

TensorFlow is an open source platform for machine learning. The GatherNd function takes arguments that determine the sizes of inputs and outputs. If the inputs given are greater than or equal to the sizes of the outputs, an out-of-bounds memory read is triggered. This issue has been patched in...

CVE-2022-35937 OOB read in `Gather_nd` op in TensorFlow Lite

TensorFlow is an open source platform for machine learning. The GatherNd function takes arguments that determine the sizes of inputs and outputs. If the inputs given are greater than or equal to the sizes of the outputs, an out-of-bounds memory read is triggered. This issue has been patched in...

CVE-2022-35939 Out of bounds write in `scatter_nd` op in TensorFlow Lite

TensorFlow is an open source platform for machine learning. The ScatterNd function takes an input argument that determines the indices of of the output tensor. An input index greater than the output tensor or less than zero will either write content at the wrong index or trigger a crash. We have...