CVE-2020-26270 | Date: (C)2020-12-11 (M)2023-12-22 |
In affected versions of TensorFlow running an LSTM/GRU model where the LSTM/GRU layer receives an input with zero-length results in a CHECK failure when using the CUDA backend. This can result in a query-of-death vulnerability, via denial of service, if users can control the input to the layer. This is fixed in versions 1.15.5, 2.0.4, 2.1.3, 2.2.2, 2.3.2, and 2.4.0.
CVSS Score and Metrics +CVSS Score and Metrics -CVSS V3 Severity: | CVSS V2 Severity: |
CVSS Score : 3.3 | CVSS Score : 2.1 |
Exploit Score: 1.8 | Exploit Score: 3.9 |
Impact Score: 1.4 | Impact Score: 2.9 |
|
CVSS V3 Metrics: | CVSS V2 Metrics: |
Attack Vector: LOCAL | Access Vector: LOCAL |
Attack Complexity: LOW | Access Complexity: LOW |
Privileges Required: LOW | Authentication: NONE |
User Interaction: NONE | Confidentiality: NONE |
Scope: UNCHANGED | Integrity: NONE |
Confidentiality: NONE | Availability: PARTIAL |
Integrity: NONE | |
Availability: LOW | |
| |