CVE-2021-29577

CVE-2021-29577 describes a heap-buffer-overflow in TensorFlow’s tf.raw_ops.AvgPool3DGrad caused by an assumption that orig_input_shape and grad have similar first/last dimensions without proper validation. Public sources (NVD, OSV, GHSA) confirm the issue and its impact as a memory corruption vul...

CVE-2021-29603

CVE-2021-29603 describes a heap out-of-bounds write in the TFLite ArgMin/ArgMax path of TensorFlow’s runtime. A specially crafted TFLite model can trigger the write when axis_value is not in [0, NumDimensions(input)], causing code to write past output_dims->data. Root cause: incorrect handling...

CVE-2021-37670

TensorFlow vulnerability CVE-2021-37670 involves a heap-based out-of-bounds read in tf.raw_ops.UpperBound (and LowerBound) caused by missing rank validation of the sorted_input argument. A patch was committed (commit 42459e4273c2e47a3232cc16c4f4fff3b3a35c38) and the fix is slated for TensorFlow 2...

CVE-2022-29195

TensorFlow CVE-2022-29195 concerns a denial-of-service via missing validation in tf.raw_ops.StagePeek. Affected are versions before patches: 2.9.0, 2.8.1, 2.7.2, and 2.6.4, which patch the issue. The vulnerability stems from StagePeek assuming index is a scalar without validating input, leading t...

CVE-2020-15266

In TensorFlow, CVE-2020-15266 affects the tf.image.crop_and_resize path: when boxes contains a very large value in versions before 2.4.0, the CPU kernel reads it as a C++ NaN and causes undefined behavior leading to a segmentation fault. The issue is fixed in commit eccb7ec454e6617738554a255d77f0...

CVE-2021-29520

TensorFlow CVE-2021-29520 concerns a heap buffer overflow in Conv3DBackprop* due to missing validation that assumes input, filter_sizes, and out_backprop have identical shapes. Multiple sources (OSV entries and GHSA advisory) corroborate the issue and patch lineage. The vulnerability affects Conv...

CVE-2021-29573

CVE-2021-29573 concerns TensorFlow. The flaw is a division-by-zero in tf.raw_ops.MaxPoolGradWithArgmax caused by the implementation not validating the batch dimension is non-zero before dividing. The issue affects TensorFlow’s MaxPoolGradWithArgmax path and is addressed in the fix that will be in...

CVE-2021-29596

CVE-2021-29596: TensorFlow/TFLite EmbeddingLookup suffers a division-by-zero in EmbeddingLookup when the first dimension of value is 0. Affected component is the TFLite operator in TensorFlow; root cause is a division by zero in embedding_lookup.cc. The documented fix is to include a patch in Ten...

CVE-2021-29601

TensorFlow/TFLite concatentation is vulnerable to an integer overflow in the TFLite kernel used for concatenation. The issue arises because TFLite uses int for tensor dimensions while TensorFlow uses int64, allowing crafted inputs where a concatenation dimension overflows an int. Public details i...

CVE-2021-29619

CVE-2021-29619 affects TensorFlow via tf.raw_ops.SparseCountSparseOutput, where passing invalid arguments (including fuzzing-derived inputs) can cause a segfault. Connected sources confirm this is a TensorFlow in-tree issue with a fix planned for TensorFlow 2.5.0 and cherry-picks in supported 2.x...

CVE-2022-23564

CVE-2022-23564 (TensorFlow) : The issue is a denial-of-service in TensorFlow caused by an invalidated CHECK assertion when decoding a resource handle tensor from protobuf, triggered by user-controlled arguments. Affected: TensorFlow (TF) processes; root cause is an assertion failure path during r...

CVE-2022-29196

CVE-2022-29196 applies to TensorFlow. Before versions 2.9.0, 2.8.1, 2.7.2, and 2.6.4, Conv3DBackpropFilterV2 does not fully validate input arguments, specifically not validating that filter_sizes is a vector. This triggers a CHECK failure and can be leveraged to cause a denial of service. The iss...

CVE-2022-29208

TensorFlow CVE-2022-29208: The tf.raw_ops.EditDistance implementation has incomplete validation, allowing crafted negative values to cause an out-of-bounds write and segmentation-fault based DoS. Affected versions are 2.6.4, 2.7.2, 2.8.1, and 2.9.0; patches exist and fixes are included in 2.9.0 (...

CVE-2022-29212

TensorFlow vulnerability CVE-2022-29212 affects TFLite model loading due to quantization scale handling. During quantization, values can have a scale > 1, but code assumed sub-unit scaling; this triggers a TFLITE_CHECK_LT assertion when QuantizeMultiplierSmallerThanOneExp is used, causing a cr...

CVE-2022-23578

TensorFlow vulnerability CVE-2022-23578 describes a memory leak: if a graph node is invalid, ImmutableExecutorState::Initialize can leak the previously allocated memory when item->kernel is reset to nullptr. The issue affects TensorFlow; the fix is planned for TensorFlow 2.8.0, with cherry-pic...

CVE-2022-29198

CVE-2022-29198 concerns TensorFlow’s tf.raw_ops.SparseTensorToCSRSparseMatrix, where input validation is incomplete for dense_shape and indices. This can trigger a CHECK failure, potentially enabling a denial of service. The issue is documented to affect TensorFlow versions prior to 2.9.0, 2.8.1,...

CVE-2022-36018

TensorFlow CVE-2022-36018 describes a denial-of-service risk in RaggedTensorToVariant when rt_nested_splits contains tensors with ranks other than one. The issue triggers a CHECK failure that can be exploited remotely to disrupt service. A fix was committed (88f93dfe691563baa4ae1e80ccde2d5c7a1438...

CVE-2022-41891

CVE-2022-41891 (TensorFlow) : A segmentation fault occurs in tf.raw_ops.TensorListConcat when element_shape is []; this can trigger a denial-of-service. A patch was committed (fc33f3dc4c14051a83eec6535b608abe1d355fde) and will be included in TensorFlow 2.11. TensorFlow 2.10.1, 2.9.3, and 2.8.4 wi...

CVE-2022-41899

CVE-2022-41899 — TensorFlow SdcaOptimizer rank check issue . The vulnerability occurs when inputs are not rank-2 and triggers a CHECK failure in SdcaOptimizer, potentially impacting availability. The root cause is a rank validation check in the optimizer. Patch available in GitHub commit 80ff197d...

CVE-2022-41902

CVE-2022-41902 in TensorFlow describes an out-of-bounds read/crash caused by MakeGrapplerFunctionItem input-size handling. A GitHub commit a65411a1d69edfb16b25907ffb8f73556ce36bb7 fixes the issue, and the fix will be released in TensorFlow 2.11.0. The same patch has been cherry-picked to TensorFl...

CVE-2022-41911

CVE-2022-41911 affects TensorFlow; root cause is an undefined char-to-bool conversion when printing a tensor, leading to sanitizer/fuzzer crashes. Patch is in GitHub commit 1be74370327 and will be included in TensorFlow 2.11.0, with backports to 2.10.1, 2.9.3, and 2.8.4. Public detail confirms im...

CVE-2021-29565

TensorFlow CVE-2021-29565 describes a null pointer dereference in tf.raw_ops.SparseFillEmptyRows due to missing validation when dense_shape is empty. Connected sources confirm the issue, including a GitHub commit (faa76f39014ed3b5e2c158593b1335522e573c7f) and notes that the fix will be included i...

CVE-2021-29578

CVE-2021-29578 : TensorFlow’s tf.raw_ops.FractionalAvgPoolGrad is vulnerable to a heap-based buffer overflow caused by a missing validation of pooling sequence arguments against the out_backprop tensor shape. Multiple connected sources corroborate the issue and provide the same root cause. The mi...

CVE-2021-29608

TensorFlow CVE-2021-29608 maps to a RaggedTensorToTensor validation flaw: input checks only ensure one tensor is non-empty, enabling potential heap out-of-bounds/NULL dereference undefined behavior in release builds. Multiple sources (NVD, OSV/GHSA advisories) describe a local-attack surface lead...

CVE-2021-41197

CVE-2021-41197 concerns TensorFlow where treating large tensor shapes can overflow int64, causing a CHECK-failure abort during shape construction (notably in operations like tf.math.segment_, SparseCwise , and depthwise-related paths). The issue is addressed by upstream fixes, with the primary pa...

CVE-2022-35935

CVE-2022-35935 affects TensorFlow via SobolSampleOp triggering a denial of service from a misvalidation that assumes input(0), input(1), and input(2) are scalar, causing a CHECK failure. The issue has been patched in the GitHub commit c65c67f88ad770662e8f191269a907bf2b94b1bf, with the fix slated ...

CVE-2022-41898

TensorFlow CVE-2022-41898 causes a crash when SparseFillEmptyRowsGrad is given empty inputs. The issue was patched in commit af4a6a3c8b95022c351edae94560acc61253a1b8 and will be included in TensorFlow 2.11; the patch will also be cherry-picked to 2.10.1, 2.9.3, and 2.8.4, which are within the sup...

CVE-2021-29554

CVE-2021-29554 affects TensorFlow: DenseCountSparseOutput can cause a denial of service via a division-by-zero runtime error. Root cause is computing a divisor from user data without validating zero, yielding num_batch_elements = 0 and a divide operation. Affected TF releases are mitigated by a f...

CVE-2021-29556

CVE-2021-29556 affects TensorFlow cores with a Denial of Service via a division-by-first-dimension in tf.raw_ops.Reverse. The vulnerability arises because N = input.dim_size(0) is used to compute cost_per_unit, enabling a user-controlled trigger (through the tensor’s first dimension) to cause a F...

CVE-2021-29584

CVE-2021-29584 and related issues affect TensorFlow where integer overflow in building tensor shapes can trigger a CHECK-fail, causing a denial-of-service via aborts on both CPU and GPU tensor shape computations (e.g., in SparseSplit and segment/shape handling). Root cause: shape construction use...

CVE-2021-29617

CVE-2021-29617 affects TensorFlow via a CHECK-fail in tf.strings.substr when given invalid arguments, causing a denial-of-service. The issue is tied to inputs to tf.strings.substr and is addressed by a fix planned for TensorFlow 2.5.0, with cherry-picks into 2.4.2, 2.3.3, 2.2.3, and 2.1.4 as thes...

CVE-2021-37662

CVE-2021-37662 stems from a TensorFlow vulnerability where a reference binding to nullptr in BoostedTreesCalculateBestGainsPerFeature and BoostedTreesCalculateBestFeatureSplitV2 may trigger undefined behavior due to input not being validated in stats_ops.cc. GitHub patches 9c87c32c7... and 429f00...

CVE-2021-37665

CVE-2021-37665 : TensorFlow vulnerable due to incomplete validation in MKL requantization (range per-channel and per-channel ops). The root cause is insufficient validation of input dimensions/arguments, allowing undefined behavior (binding a null pointer, heap out-of-bounds access). A patch was ...

CVE-2022-29210

CVE-2022-29210 affects TensorFlow 2.8.0, where TensorKey's hash using AllocatedBytes() and tensor.data() can cause a heap-buffer/ASAN-related issue leading to a denial of service. The issue is patched in TensorFlow 2.9.0 and also back-ported to 2.8.1. IBM Watson Discovery bulletin and OSV entries...

CVE-2022-35991

TensorFlow CVE-2022-35991 affects TensorListScatter and TensorListScatterV2 when element_shape has rank greater than one, triggering a CHECK failure that can lead to a denial of service. The issue is documented in OSV entries (BIT-TENSORFLOW-2022-35991) and related advisories, which confirm a pat...

CVE-2022-36027

TensorFlow's CVE-2022-36027 affects the TFLite/TensorFlow converter path for per-channel quantized transposed convolutions, where converting these ops can segfault and crash the Python process. The issue has been patched in commit aa0b852a4588cea4d36b74feb05d93055540b450, with the fix slated for ...

CVE-2021-29526

Summary: CVE-2021-29526 affects TensorFlow’s Conv2D path, where the division by a caller-controlled quantity in tf.raw_ops.Conv2D can trigger a division by zero. This vulnerability is supported by multiple sources in connected documents (OSV entries and NVD/CVE references) that describe the under...

CVE-2021-29555

TensorFlow CVE-2021-29555 describes a denial-of-service vulnerability in tf.raw_ops.FusedBatchNorm caused by a division operation based on the last tensor dimension, which can be triggered by user-provided input. The issue affects the FusedBatchNorm path and has been patched; TensorFlow 2.5.0 wil...

CVE-2021-37635

CVE-2021-37635 affects TensorFlow: heap out-of-bounds access in sparse reduction operations due to missing validation of reduction groups/indices. Patch committed (87158f43f05f2720a374f3e6d22a7aaa3a33f750) and fixes planned for TensorFlow 2.6.0, with cherry-picks for 2.5.1, 2.4.3, and 2.3.4. Publ...

CVE-2021-37669

TensorFlow CVE-2021-37669 describes a denial-of-service in tf.raw_ops.NonMaxSuppressionV5 (and CombinedNonMaxSuppression) caused by an integer-to-unsigned conversion when resizing a std::vector with a user-controlled output_size. Negative output_size leads to a crash due to implicit conversion to...

CVE-2021-37692

CVE-2021-37692 affects TensorFlow and centers on a segfault in string tensor deallocation during garbage collection when the encoding of a string tensor fails (e.g., mismatched dimensions). The root cause is an assumption that encoding succeeded, leading to use of the finalizer of the tensor with...

CVE-2022-21738

TensorFlow CVE-2022-21738 describes an integer overflow in SparseCountSparseOutput (kernels/count_ops.cc) that can crash a TensorFlow process when its result is used for memory allocation. The vulnerability affects TensorFlow releases including 2.5.3, 2.6.3, and 2.7.1, with a fix planned for Tens...

CVE-2022-23575

TensorFlow vulnerability CVE-2022-23575 arises from an integer overflow in OpLevelCostEstimator::CalculateTensorSize when processing an operation with a very large tensor. The issue affects TensorFlow and was mitigated by a patch in TensorFlow 2.8.0, with cherry-picks to 2.7.1, 2.6.3, and 2.5.3 (...

CVE-2022-23590

CVE-2022-23590 affects TensorFlow. A maliciously altered GraphDef from a SavedModel can trigger a crash by calling ValueOrDie on an error StatusOr, crashing the TensorFlow process. Patches exist in GitHub commits and will be included in TensorFlow 2.8.0 and 2.7.1; remediation in practice is to up...

CVE-2022-35974

Summary: CVE-2022-35974 affects TensorFlow via QuantizeDownAndShrinkRange when given nonscalar input_min/input_max, causing a segfault and a potential denial of service. The issue has been patched in commit 73ad1815ebcfeb7c051f9c2f7ab5024380ca8613 and the fix will be included in TensorFlow 2.10.0...

CVE-2022-35998

CVE-2022-35998 affects TensorFlow. When EmptyTensorList receives an input element_shape with more than one dimension, a CHECK failure can be triggered, potentially enabling a denial of service. The issue is fixed in commit c8ba76d48567aed347508e0552a257641931024d and will be included in TensorFlo...

CVE-2022-41908

TensorFlow CVE-2022-41908: CHECK fail in tf.raw_ops.PyFunc triggered by non-UTF-8 input tokens. Patch committed (9f03a9d3bafe902c1e6beb105b2f24172f238645); fix slated for TensorFlow 2.11 with cherry-picks to 2.10.1, 2.9.3, and 2.8.4. No exploit details provided in the documents.

CVE-2019-16778

CVE-2019-16778 : In TensorFlow before 1.15, a heap-based buffer overflow can occur in UnsortedSegmentSum when the Index template argument is int32. Data_size and num_segments are truncated from int64 to int32, potentially producing negative numbers and out-of-bounds memory access. The issue is de...

CVE-2021-29530

CVE-2021-29530 concerns TensorFlow’s SparseMatrixSparseCholesky: passing an invalid permutation could trigger a null pointer dereference due to incomplete input validation in the underlying code, allowing continuation after a failed check. Public sources (NVD, OSV, GHSA) describe the issue in Ten...

CVE-2021-29535

CVE-2021-29535 affects TensorFlow’s QuantizedMul implementation. The vulnerability arises when four quantization thresholds are passed as inputs and may be empty tensors, causing a heap buffer overflow by accessing .flat()(0) on an empty buffer. This is rooted in the code path in quantized_mul_op...