Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
The authors introduce SecLens-R, a multi-stakeholder evaluation framework for LLM-based vulnerability detection, structured around 35 dimensions and 5 role-specific weighting profiles (CISO, Chief AI Officer, Security Researcher, Head of Engineering, and AI-as-Actor). They evaluate 12 frontier models on a dataset of 406 tasks across 10 programming languages and 8 vulnerability categories, in both Code-in-Prompt and Tool-Use settings. Results reveal significant performance variation across stakeholder perspectives, with Decision Scores differing by up to 31 points for the same model, highlighting the limitations of single aggregated metrics.
LLMs can ace security vulnerability detection for engineering leaders but completely fail from a CISO's perspective, proving that a one-size-fits-all evaluation metric is dangerously misleading.
Existing benchmarks for LLM-based vulnerability detection compress model performance into a single metric, which fails to reflect the distinct priorities of different stakeholders. For example, a CISO may emphasize high recall of critical vulnerabilities, an engineering leader may prioritize minimizing false positives, and an AI officer may balance capability against cost. To address this limitation, we introduce SecLens-R, a multi-stakeholder evaluation framework structured around 35 shared dimensions grouped into 7 measurement categories. The framework defines five role-specific weighting profiles: CISO, Chief AI Officer, Security Researcher, Head of Engineering, and AI-as-Actor. Each profile selects 12 to 16 dimensions with weights summing to 80, yielding a composite Decision Score between 0 and 100. We apply SecLens-R to evaluate 12 frontier models on a dataset of 406 tasks derived from 93 open-source projects, covering 10 programming languages and 8 OWASP-aligned vulnerability categories. Evaluations are conducted across two settings: Code-in-Prompt (CIP) and Tool-Use (TU). Results show substantial variation across stakeholder perspectives, with Decision Scores differing by as much as 31 points for the same model. For instance, Qwen3-Coder achieves an A (76.3) under the Head of Engineering profile but a D (45.2) under the CISO profile, while GPT-5.4 shows a similar disparity. These findings demonstrate that vulnerability detection is inherently a multi-objective problem and that stakeholder-aware evaluation provides insights that single aggregated metrics obscure.