CVE-2025-46153: PyTorch before 3.7.0 has a bernoulli_p decompose function in decompositions.py even though it lacks full consistency wit
mediumvulnerability
security
Summary
PyTorch versions before 3.7.0 have a bug in the bernoulli_p decompose function (a mathematical operation used in the dropout layers) that doesn't work the same way as the main CPU implementation, causing problems with nn.Dropout1d, nn.Dropout2d, and nn.Dropout3d when fallback_random=True (a setting that uses random number generation as a backup method).
Vulnerability Details
CVSS Score
5.3(medium)
EPSS (30-day exploit probability)
EPSS: 0.1%
Classification
Attack SophisticationModerate
Impact (CIA+S)
integrity
AI Component TargetedFramework
Taxonomy References
CWE (Weakness Type)
Affected Vendors
Monthly digest — independent AI security research
Original source: https://nvd.nist.gov/vuln/detail/CVE-2025-46153
First tracked: February 15, 2026 at 08:37 PM
Classified by LLM (prompt v3) · confidence: 85%