PT-2025-39381 · Pytorch+1 · Pytorch+1

Published 2024-11-26 · Updated 2025-10-05 · CVE-2025-46150

Positive Technologies (AI Summary)

CVSS v3.1

5.3

Medium

Base vector

AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N

Name of the Vulnerable Software and Affected Versions PyTorch versions prior to 2.7.0

Description When torch.compile is used, the FractionalMaxPool2d function exhibits inconsistent results. This issue affects the accuracy of computations involving this specific function within the PyTorch framework. The FractionalMaxPool2d function is used for performing fractional max pooling operations, a common technique in deep learning models. The use of torch.compile introduces a performance optimization feature that, in this case, leads to the observed inconsistency.

Recommendations Update to PyTorch version 2.7.0 or later.

Fix

Buffer Overflow

Weakness Enumeration

CWE-119

Related Identifiers

BDU:2025-13136

BIT-PYTORCH-2025-46150

CVE-2025-46150

PYSEC-2025-200

Affected Products

Debian

Pytorch

References · 17

• https://github.com/pytorch/pytorch/pull/144395 · Patch
• https://ubuntu.com/security/CVE-2025-46150 · Vendor Advisory
• https://osv.dev/vulnerability/BIT-pytorch-2025-46150 · Vendor Advisory
• https://bdu.fstec.ru/vul/2025-13136 · Security Note
• https://osv.dev/vulnerability/DEBIAN-CVE-2025-46150 · Vendor Advisory
• https://security-tracker.debian.org/tracker/CVE-2025-46150 · Vendor Advisory
• https://nvd.nist.gov/vuln/detail/CVE-2025-46150 · Security Note
• https://osv.dev/vulnerability/CVE-2025-46150 · Vendor Advisory
• https://osv.dev/vulnerability/UBUNTU-CVE-2025-46150 · Vendor Advisory
• https://cve.org/CVERecord?id=CVE-2025-46150 · Security Note
• https://osv.dev/vulnerability/PYSEC-2025-200 · Vendor Advisory
• https://github.com/pytorch/pytorch/issues/141538#issuecomment-2537424658 · Note
• https://github.com/pytorch/pytorch/issues/141538 · Note
• https://t.me/cveNotify/137350 · Telegram Post
• https://twitter.com/VulmonFeeds/status/1971332992514637976 · Twitter Post