CVE detail

CVE-2020-28975 — CVE-2020-28975

Published 2020-11-21 · Modified 2026-06-17 · Vendor scikit-learn · Product scikit-learn · Source nvd

HIGH

severity

CVSS-derived band

7.5

CVSS v3

0–10 scale

0.0346

EPSS probability

exploitation probability, 30d

88.0%

EPSS percentile

percentile vs all CVEs

NOT LISTED

CISA KEV

known exploited catalog

Description

svm_predict_values in svm.cpp in Libsvm v324, as used in scikit-learn 0.23.2 and other products, allows attackers to cause a denial of service (segmentation fault) via a crafted model SVM (introduced via pickle, json, or any other model permanence standard) with a large value in the _n_support array. NOTE: the scikit-learn vendor's position is that the behavior can only occur if the library's API is violated by an application that changes a private attribute.

Remediation

No vendor-published fix data in our corpus for this CVE. Check the references below or the vendor's PSIRT / security advisories page.

References

cvedb.io · NVD · CISA KEV · FIRST EPSS · vendor advisories (CVE Program List v5). Informational only, no warranty — verify every remediation against the vendor advisory before acting on it. This product uses data from the NVD API but is not endorsed or certified by the NVD, CISA, FIRST.org or any vendor named. CVE® is a registered trademark of The MITRE Corporation.