Jul 2026· AIware· pp. 388-396· 2 citations· 29 references
Computer Science
TL;DR
This paper introduces SecVulEval, a context-aware benchmark designed to evaluate LLMs on vulnerability detection with rich contextual information, and believes it can serve as a foundation for advancing context-aware vulnerability detection with LLMs.
Abstract
Large Language Models (LLMs) show promise for vulnerability detection, but their evaluation is limited by the lack of high-quality benchmarks. Most existing datasets rely on coarse function-level labels, overlook fine-grained vulnerability patterns, and lack critical program context such as data/control dependencies. They also suffer from data quality issues, including mislabeling and duplication, leading to unreliable evaluation and limited real-world relevance. To address these limitations, this paper introduces SecVulEval, a context-aware benchmark designed to evaluate LLMs on vulnerability detection with rich contextual information. SecVulEval focuses on real-world C/C++ vulnerabilities at the statement level. This granularity enables more precise evaluation of a model’s ability to localize and understand vulnerabilities, beyond simple binary classification at the function level. By incorporating rich contextual information, SecVulEval sets a new standard for benchmarking vulnerability detection in realistic software development scenarios. This benchmark includes 25,440 function samples covering 5,867 unique CVEs in C/C++ projects from 1999 to 2024. We evaluated state-of-the-art LLMs in both standalone and multi-agent settings. Results on our dataset indicate that current models remain far from accurately identifying vulnerable statements within a given function, although agent-based approaches provide modest but promising improvements. The best-performing Claude-3.7-Sonnet-driven agent achieves an F1-score of 23.83% for vulnerable statement detection. We believe this benchmark can serve as a foundation for advancing context-aware vulnerability detection with LLMs.
VulnGym is a real-world repository-level benchmark for evaluating vulnerability detection by coding agents that aligns reviewed GitHub advisories with their corresponding vulnerable version repositories and defines an end-to-end detection task and three oracle-based subtasks to jointly evaluate vulnerability detection and diagnose limitations in code localization and evidence construction.
Kexing Ji, Jiachen Liu, En-Ze Hu et al.· 2 citations
VICBench enables robust evaluation of vulnerability detection approaches and shows that state-of-the-art algorithms V-SZZ and LLM4SZZ achieve only 33.3%-40.1% F1, confirming that using existing approaches still entails significant manual effort.
Jin Lu, Xuening Han, Yan Zhong et al.· 0 citations
With more than 20,000 Common Vulnerabilities and Exposures (CVEs) reported annually, software vulnerabilities represent a critical cybersecurity challenge. This volume has intensified the demand for automated detection and analysis, motivating the integration of large language models (LLMs) for such tasks. However, existing vulnerability benchmarks are not suitable for evaluating LLMs' capabilities in vulnerability assessment, as most of them 1) rely on narrow data sources, 2) lack deep context, and 3) focus on single-turn Q&A rather than realistic, multi-stage analyst workflows. To address this gap, we introduce OmniVul, a comprehensive multi-turn benchmark for LLM-based vulnerability assessment. OmniVul comprises 2,000 CVEs with question–answer pairs spanning 23 attributes, including detection, code localization, root cause analysis, and patch suggestion. We employ an automated workflow to aggregate multi-source data via Retrieval-Augmented Generation (RAG), ensuring quality through LLM-as-a-Judge filtering and conformal prediction calibrated by human expert annotations. An evaluation of five state-of-the-art LLMs on OmniVul reveals distinct performance gaps, with top-1 accuracy remaining below 50% on average for vulnerable code detection and CVE identification. Our evaluation also demonstrates that current models lack critical reasoning capabilities for reliable vulnerability assessment. These results highlight the importance of OmniVul for advancing research in evaluating and fine-tuning LLMs for vulnerability assessment.
Vishnu Teja Kandalam, Viet Duong, Xiaochang Li et al.· Proceedings of the 32nd ACM...· 0 citations
: Large language models are increasingly being deployed in safety-critical domains, yet remain vulnerable to jailbreak attacks that circumvent safety alignments. This systematic review synthesizes empirical jailbreak research published between 2024 and 2025, using a PRISMA-guided search protocol, followed by BERTopic-based topic modeling. The analysis identifies eight main jailbreak categories: optimization-based, ge-netic/evolutionary, iterative refinement, semantic/persuasion-based, decomposition, context/generation-level, visual/encoding and fuzzing attacks, and characterizes their effectiveness, efficiency, and transferability across open-source and proprietary models, including Llama-2/3, Vicuna, GPT-3.5/4, Claude, Gemini, and DeepSeek-V3. Results show that simple configuration and context-level attacks can match the near-perfect attack success rates of sophisticated white-box optimization methods on models such as Llama-2, while requiring far fewer queries and no parameter access, highlighting a gap between research focus and practical threat severity. The review further identifies five recurring vulnerability mechanisms: representation-level gaps, execution-priority manipulation, semantic fragmentation, gradient-space exploitation and persuasion susceptibility, and documents family-specific vulnerability patterns, with open-source Llama-based models consistently more exposed than safety-enhanced architectures such as Claude. Diverse methods, uneven focus on models and publication bias limit how broadly results apply. Nonetheless, the review reveals that weaknesses in safety alignment persist across successive LLM generations, urging that effective defenses must address all eight attack categories rather than isolated techniques.
Meda Račaitytė, Hélder Bastos, R. Ribeiro et al.· International Conference on...· 0 citations
Vulnerability detection via static analysis traditionally relies on security experts encoding insecure coding patterns into algorithmic rules. However, this approach often focuses on syntactic patterns and overlooks deeper semantic information in the code, such as the meanings of variable and function names. As software systems grow more complex, modeling vulnerabilities using only syntactic rules becomes increasingly challenging. In this paper, we propose a semantics-aware approach to detecting software vulnerabilities. We present SETYPE, a semantics-aware type system that can be derived directly from source code based solely on the meanings of symbols and expressions in natural language. In the SETYPE type system, both type inference and checking are performed by Large Language Models (LLMs), and a failed type check indicates a potential vulnerability. We prototype PYSETYPE to demonstrate the feasibility of SETYPE for detecting vulnerabilities in Python web applications. Our evaluation on real-world applications achieves 87% detection precision and 88% detection accuracy. Using PYSETYPE, we identified 15 potential zero-day vulnerabilities, nine of which were confirmed by developers.
Detecting known vulnerabilities in binary functions requires associating each function with relevant Common Vulnerabilities and Exposures (CVE) knowledge and determining the function’s state with respect to a specified CVE. Compilation obscures source-level semantics, while security patches often modify only a local check or control-flow condition. We present BinVulRAG, a function-level assessment framework that retrieves provenance-aware knowledge units rather than plain documents or code fragments. Each unit integrates vulnerability conditions, vulnerability and repair behavior, binary-difference evidence, and provenance metadata. During inference, given a target CVE, BinVulRAG retrieves relevant knowledge units and assesses vulnerability and repair behavior. The confidence-gating mechanism then assigns the final labels: vulnerable, non-vulnerable, or unknown. On 380 vulnerable–patched function pairs, BinVulRAG achieved F1 scores of 66.2% (vulnerable-class) and 59.1% (non-vulnerable-class), and a pair accuracy of 26.1%. These three metrics exceeded the corresponding best baseline values by 4.2, 11.2, and 12.2 percentage points, respectively. The ablation results show different precision–recall profiles across evidence configurations, indicating complementary roles for paired-binary evidence and structured vulnerability knowledge. These findings support grounding retrieved vulnerability knowledge in binary-observable evidence when distinguishing vulnerable functions from patched functions.
Ye Tian, Qian Xiao, Hongmei Wei et al.· Electronics· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
