Backend
CyBiasBench: Benchmarking Bias in LLM Agents for Cyber-Attack Scenarios
arXiv cs.CR0 views
arXiv:2605.07830v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly deployed as autonomous agents in offensive cybersecurity. In this paper, we reveal an interesting phenomenon: different agents exhibit distinct attack patterns. Specifically, agents exhibit distinct attack-selection biases, differing in both the families they explore and the concentration of their efforts. These patterns remain relatively stable across prompt variations. To systematically quantify this behavior, we introduce CyBiasBench, a comprehensive 630-session benchmark that evaluates five agents on three targets and four prompt conditions with ten attack families. We identify explicit bias across agents, with different dominant attack families and varying entropy levels in their attack-family allocation distributions. Such bias characterizes agent behavior within the evaluated environments and provides information beyond aggregate attack success rates. Furthermore, our experiments reveal a bias momentum effect, where agents resist explicit steering toward attack families that conflict with their bias. This forced distribution shift does not yield measurable improvements in attack performance. To ensure reproducibility and facilitate future research, we release an interactive result dashboard at https://trustworthyai.co.kr/CyBiasBench/ and a reproducibility artifact with aggregated session-level statistics and full evaluation scripts at https://github.com/Harry24k/CyBiasBench.
Source:https://arxiv.org/abs/2605.07830PreviousTowards Reliable Local Security Agents: Verifiable Post-Training for Linux Privilege EscalationNextIH-Benchmark: A Conflict-Centered Benchmark for Instruction-Hierarchy Robustness in Large Language Models
Related posts
SQL Server migration verification: matching row counts are only the start
0dev.to Database

The Claim Row Stopped the Duplicate. My Agent Sent the Payment Twice Anyway.
0dev.to Database

SeaweedFS 4.54: S3 Storage, Iceberg Tables and Vector Buckets
0dev.to Database
Where Is the Cost of Third-Party API Routers in Agentic Software Development?
0arXiv cs.SE
RyBench: Runtime Verification Benchmarking Framework over Publish/Subscribe Networks
0arXiv cs.SE
When the World Is the Test Suite: Agentic Intelligence for the Physical World
0arXiv cs.SE