1373 matches found
Improper Validation of Specified Quantity in Input
Overview Affected versions of this package are vulnerable to Improper Validation of Specified Quantity in Input in the ModularIndexing function when Inductor config is set to constantandindexpropagation=False. An attacker can cause incorrect computation results by supplying crafted input data...
Use of Out-of-range Pointer Offset
Overview Affected versions of this package are vulnerable to Use of Out-of-range Pointer Offset via the fractionalmaxpool function in torch.nn.FractionalMaxPool2d component when used with torch.compile. An attacker can cause inconsistent computational results by exploiting this behavior,...
Use of Out-of-range Pointer Offset
Overview torch is a Tensors and Dynamic neural networks in Python with strong GPU acceleration Affected versions of this package are vulnerable to Use of Out-of-range Pointer Offset via the fractionalmaxpool function in torch.nn.FractionalMaxPool2d component when used with torch.compile. An...
Inefficient CPU Computation
Overview Affected versions of this package are vulnerable to Inefficient CPU Computation due to inconsistent behavior in the bernoullip function when used in RNG nn.Dropout1d, nn.Dropout2d, and nn.Dropout3d. An attacker can cause unintended or incorrect dropout behavior in neural network layers b...
Reachable Assertion
Overview Affected versions of this package are vulnerable to Reachable Assertion via the nn.Fold function when used with inductor. An attacker can cause the application to terminate unexpectedly by triggering a buffer overflow error. Remediation Upgrade pytorch/pytorch to version 2.7.0-rc1 or...
PyTorch 安全漏洞
PyTorch is a Python package open-sourced by PyTorch. An unspecified vulnerability exists in the PyTorch torch.linalg.lu component, which can be exploited by an attacker to cause a denial of service...
CVE-2025-55557
CVE-2025-55557 : A NameError occurs in PyTorch v2.7.0 when a model uses torch.cummin and is compiled by Inductor, leading to a Denial of Service. The initial description provides the affected components and the root cause, but no public remediation or exploitation details are documented in the su...
PT-2025-39382
Name of the Vulnerable Software and Affected Versions PyTorch versions prior to 2.7.0 Description The bitwise right shift function in PyTorch produces incorrect output when the other argument has certain out-of-bounds values. This issue affects versions prior to 2.7.0. Recommendations Update to...
PyTorch 安全漏洞
PyTorch is a Python package open-sourced by PyTorch. PyTorch suffers from a security vulnerability that stems from the bitwiserightshift function mishandling the boundary value of the OTHER parameter, which can be exploited by an attacker to cause an output error...
CVE-2025-46150
In PyTorch before 2.7.0, when torch.compile is used, FractionalMaxPool2d has inconsistent results...
CVE-2025-55553
A syntax error in the component proxytensor.py of pytorch v2.7.0 allows attackers to cause a Denial of Service DoS...
CVE-2025-55552
pytorch v2.8.0 was discovered to display unexpected behavior when the components torch.rot90 and torch.randnlike are used together...
PT-2025-39416
Name of the Vulnerable Software and Affected Versions pytorch version 2.7.0 Description A buffer overflow can occur when a PyTorch model includes torch.nn.Conv2d, torch.nn.functional.hardshrink, and torch.Tensor.view-torch.mv and is compiled using Inductor. This can lead to a Denial of Service Do...
CVE-2025-46150
In PyTorch versions prior to 2.7.0 , using torch.compile with the FractionalMaxPool2d component leads to inconsistent results . While one source suggests an information disclosure vulnerability, specific technical details regarding the root cause or exploitation are not fully provided. Additional...
CVE-2025-55558
A buffer overflow vulnerability exists in pytorch v2.7.0 caused by a boundary error during model compilation when handling untrusted input. The flaw is triggered when a model contains a specific combination of torch.nn.Conv2d , torch.nn.functional.hardshrink , and torch.Tensor.view-torch.mv() and...
CVE-2025-55553: Uncaught Exception
A syntax error in the component proxytensor.py of pytorch v2.7.0 allows attackers to cause a Denial of Service DoS...
CVE-2025-55557: Uncaught Exception
A Name Error occurs in pytorch v2.7.0 when a PyTorch model consists of torch.cummin and is compiled by Inductor, leading to a Denial of Service DoS...
CVE-2025-55551: Uncontrolled Resource Consumption
An issue in the component torch.linalg.lu of pytorch v2.8.0 allows attackers to cause a Denial of Service DoS when performing a slice operation...
