CVE-Bench
PulseAugur coverage of CVE-Bench — every cluster mentioning CVE-Bench across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
Survey details AI slop and hallucinations in cybersecurity vulnerability assessment
A new survey paper published on arXiv addresses the issue of "AI slop" and hallucinations in AI-driven vulnerability assessment. The paper identifies a core problem: the discrepancy between the deductive reasoning of hu…
-
OrcaRouter releases OrcaCyber Zero 1.5 with 1M context for cybersecurity
OrcaRouter has launched OrcaCyber Zero 1.5, a new cybersecurity model designed for vulnerability research and exploit development. This model boasts a 1 million token context window and claims top performance on benchma…
-
AI agents revolutionize security testing, discovering thousands of vulnerabilities
AI-driven agents are revolutionizing security testing, moving from auxiliary tools to autonomous hunters. Systems like XBOW have topped bug bounty leaderboards, discovering thousands of vulnerabilities at a fraction of …
-
Fireworks AI highlights FailSafe SWARM's #1 CVE-Bench score
Fireworks AI's inference infrastructure platform has highlighted FailSafe SWARM's achievement of ranking first on the CVE-Bench benchmark. FailSafe SWARM reportedly scored 62.5%, which is nearly double the performance o…
-
LLM agents struggle to patch security bugs, leaving vulnerabilities open
A new benchmark, CVE-Bench, was developed to evaluate LLM agents' ability to patch security vulnerabilities in Python projects. Across 18 projects and 20 real-world CVEs, the best performing models achieved only a 50% s…