Local code execution via unsafe scaler deserialization
Published Jan 20, 2026 · Updated Jan 20, 2026
Code injection in NVIDIA Merlin Transformers4Rec before commit 27ddd49 allows local users to execute code through a crafted scaler.pt file. Trainer.load_model_trainer_states_from_checkpoint passes scaler.pt to torch.load, which deserializes attacker-controlled pickle data without restricting executable objects. Exploitation requires local low-privileged access to place or replace the checkpoint file, after which code runs with the loading process's privileges and can read or alter its data.
Summary
What happened
Code injection in NVIDIA Merlin Transformers4Rec before commit 27ddd49 allows local users to execute code through a crafted scaler.pt file. Trainer.load_model_trainer_states_from_checkpoint passes scaler.pt to torch.load, which deserializes attacker-controlled pickle data without restricting executable objects. Exploitation requires local low-privileged access to place or replace the checkpoint file, after which code runs with the loading process's privileges and can read or alter its data.
The record
- CVE
- CVE-2025-33233
- Published
- Jan 20, 2026
- Updated
- Jan 20, 2026
- Vendor
- DPDK Project
- Product
- Transformers4Rec
- Classifications
- CWE-94
- Attack vector
- local
- Privileges
- authenticated
Timeline
How it unfolded
- Jan 20, 2026CVE publishedPublication date reported by the CVE source.
- Jan 20, 2026Record 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=All code branches that do not include commit 27ddd49
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
No public exploit references are available in this record.
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