Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Vulnerability Detection research benchmarks – Page 4 · SOTA2 Research

Vulnerability Detection

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

unified C/C++ (evaluation set)

3,681TP (Clean)

5

Feb 26, 2026

64.94Accuracy

5

Feb 26, 2026

0.738Accuracy

5

Feb 26, 2026

PreciseBugs full code completions

59F1 Score

5

Feb 27, 2026

PrimeVul Balanced

52.75Precision

4

Jun 30, 2026

62Precision

4

Jun 30, 2026

CyberGym 100 projects

175Vulnerabilities Detected Count

4

Jun 23, 2026

Recent memory-safety related CVEs (Fresh-target evaluation)

31Vulnerabilities Found

4

Jun 23, 2026

OpenClawBench held-out set n=217 (test)

CodeQL + CLAWAUDIT

75.1Recall

4

Jun 23, 2026

OpenClawBench rule-derivation set n=229 (train)

CodeQL + CLAWAUDIT

79Recall

4

Jun 23, 2026

20Issues Detected

4

Jun 11, 2026

58Issues Detected

4

Jun 11, 2026

146Issues Detected

4

Jun 11, 2026

Real-world Go bugs dataset 8 vulnerability cases

0.18Average Detection Time

4

May 6, 2026

1,050True Positives (TP)

4

Apr 30, 2026

780True Positives (TP)

4

Apr 30, 2026

0Vulnerability Count

4

Mar 9, 2026

4Unique Vulnerabilities Found

4

Mar 9, 2026

CWE-Bench-Java 120 vulnerabilities (full)

78Detected Count

4

Mar 4, 2026

91.2Precision

4

Feb 26, 2026

Six real-world firmware benchmarks (OpenSSL, HTSlib, IEC61850, MXML, WolfBoot, WolfMQTT)

118Detected CVEs

4

Feb 26, 2026

3Time (hours)

4

Feb 26, 2026

Top100 1.0 (test)

100Overall Success Rate

4

Feb 26, 2026

Vulnerability Detection synthetic dataset 1.0 (test)

Multi-turn C-ing Clearly

12.78F1 Score

4

Feb 26, 2026

SkillVetBench Memory Poisoning

1Malicious Verdict Count

3

Jun 2, 2026

Page 4 of 6

Popular datasets

PrimeVul SkillVetBench Real-world Go bugs dataset Bigvul SARD

Follow for updates

Join SOTA2 Contact