Silent incorrectness via XLA-compiled Embedding
Published Sep 25, 2025 · Updated Sep 25, 2025
Incorrect calculation in TensorFlow 2.18.0 allows remote attackers to corrupt model output via an XLA-compiled Embedding layer. With input_dim set to 1 and an out-of-range index of 1, the JIT-compiled path returns a random tensor instead of the zero produced without XLA. Exploitation requires an application to accept an attacker-selected model or index and enable XLA JIT compilation; the corrupted output can silently drive incorrect decisions.
Summary
What happened
Incorrect calculation in TensorFlow 2.18.0 allows remote attackers to corrupt model output via an XLA-compiled Embedding layer. With input_dim set to 1 and an out-of-range index of 1, the JIT-compiled path returns a random tensor instead of the zero produced without XLA. Exploitation requires an application to accept an attacker-selected model or index and enable XLA JIT compilation; the corrupted output can silently drive incorrect decisions.
The record
- CVE
- CVE-2025-55556
- Published
- Sep 25, 2025
- Updated
- Sep 25, 2025
- Vendor
- Unknown vendor
- Product
- Unknown product
- Classifications
- CWE-506, T1565
- Attack vector
- network
- Privileges
- unauthenticated
Timeline
How it unfolded
- Sep 25, 2025CVE publishedPublication date reported by the CVE source.
- Sep 25, 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?
What conditions does exploitation require?
What is affected?
Affected products and versions are unavailable in this record.
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.
Public exploit references
- XLA Embedding incorrect-output 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