Process crash via invalid LSTM gate weights
Published Mar 31, 2025 · Updated Mar 31, 2025
Memory corruption in torch.lstm_cell in PyTorch 2.6.0 allows local users to crash the calling process via malformed gate weights. The native RNN cell path accepts weight_ih and weight_hh tensors whose first dimensions do not equal four times the hidden size, then uses them without rejecting the mismatch. Execution requires local low-privileged access to run Python code; the demonstrated consequence is a segmentation fault in that process.
Summary
What happened
Memory corruption in torch.lstm_cell in PyTorch 2.6.0 allows local users to crash the calling process via malformed gate weights. The native RNN cell path accepts weight_ih and weight_hh tensors whose first dimensions do not equal four times the hidden size, then uses them without rejecting the mismatch. Execution requires local low-privileged access to run Python code; the demonstrated consequence is a segmentation fault in that process.
The record
- CVE
- CVE-2025-3001
- Published
- Mar 31, 2025
- Updated
- Mar 31, 2025
- Vendor
- PyTorch
- Product
- PyTorch
- Classifications
- CWE-119
- Attack vector
- local
- Privileges
- authenticated
Timeline
How it unfolded
- Mar 31, 2025CVE publishedPublication date reported by the CVE source.
- Mar 31, 2025Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=2.6.0
What conditions does exploitation require?
What is affected?
Published CVSS scores
CVSS describes severity. EPSS estimates exploitation probability.
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
- torch.lstm_cell segmentation-fault reproducerproof of concept · demonstrated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo