Incorrect inference via dtype-omitting cache key
Published Jun 5, 2026 · Updated Jun 5, 2026
Cache-key collision in ONNX-MLIR 0.5.0 allows local users to cause silently incorrect inference results through placeholder dtype changes. In backend.py, generate_hash_key records a placeholder's shape but omits its tensor dtype when default lightweight hashing builds the compiled-model cache key. A same-process, same-shape model with a different dtype can therefore reuse the first model's compiled shared object, corrupting inference output without an error.
Summary
What happened
Cache-key collision in ONNX-MLIR 0.5.0 allows local users to cause silently incorrect inference results through placeholder dtype changes. In backend.py, generate_hash_key records a placeholder's shape but omits its tensor dtype when default lightweight hashing builds the compiled-model cache key. A same-process, same-shape model with a different dtype can therefore reuse the first model's compiled shared object, corrupting inference output without an error.
The record
- CVE
- CVE-2026-11329
- Published
- Jun 5, 2026
- Updated
- Jun 5, 2026
- Vendor
- ONNX
- Product
- ONNX-MLIR
- Classifications
- CWE-327, CWE-328
- Attack vector
- local
- Privileges
- authenticated
Timeline
How it unfolded
- Jun 5, 2026CVE publishedPublication date reported by the CVE source.
- Jun 5, 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=0.5.0
What conditions does exploitation require?
What is affected?
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