arXiv AI

Benchmark Scores Are Pipeline-Dependent: A Reliability Audit of Cybersecurity LLM Benchmarks

The paper examines how the scores of cybersecurity large language model (LLM) benchmarks vary depending on the evaluation pipeline used. By auditing eight benchmarks across ten different LLMs, the authors uncover 15 systematic failure modes and demonstrate that a single pipeline choice can shift a model’s score by over 80 percentage points, significantly altering rankings. They also show that even semantically similar tasks can produce different model rankings due to incompatible evaluation conventions, and that standardizing pipelines can move most models by at least three ranks on at least one benchmark.

arXiv AI

Aug 19

Benchmarking the Benchmarks: Evaluating Automated Safety Benchmarks for Small Language Models

The paper investigates whether existing AI safety benchmarks, designed for large language models, are suitable for evaluating small language models (SLMs). By testing five benchmark suites on 26 open‑source SLMs with a unified scoring rubric, the authors find that ambiguous judgments dominate, especially for complex prompts and certain architectures. This ambiguity, linked to factors like lexical density and output perplexity, undermines the reliability of aggregate leaderboards and reveals a confound between model capability and perceived safety.

By Nyamtulla Shaik, Fengjun Li, Bo Luo

llms benchmarks safety

arXiv AI

Jul 2

Toward Cybersecurity-Expert Small Language Models

arXiv:2510. 14113v2 Announce Type: replace-cross Abstract: Large language models (LLMs) are transforming everyday applications, yet deployment in cybersecurity lags due to a lack of high-quality, domain-specific models and training datasets.

By Matan Levi, Daniel Ohayon, Ariel Blobstein, Ravid Sagi, Ian Molloy, Yair Allouche

llms benchmarks

arXiv Computation and Language

Sep 23

SafetyFlow: An Agent-Flow System for Automated LLM Safety Benchmarking

arXiv:2508.15526v2 Announce Type: replace Abstract: The rapid proliferation of large language models (LLMs) has intensified the requirement for reliable safety evaluation to uncover model vulnerabili...

By Xiangyang Zhu, Yuan Tian, Chunyi Li, Kaiwei Zhang, Wei Sun, Guangtao Zhai

llms agents benchmarks safety

arXiv AI

Jun 4

CyberGym-E2E: Scalable Real-World Benchmark for AI Agents' End-to-End Cybersecurity Capabilities

arXiv:2606. 04460v1 Announce Type: cross Abstract: AI has the potential to transform cybersecurity by enabling systems that can autonomously detect, analyze, and remediate software vulnerabilities.

By Tianneng Shi, Robin Rheem, Dongwei Jiang, Mona Wang, Francisco De La Riega, Zhun Wang, Jingzhi Jiang, Alexander Cheung, Sean Tai, Jonah Cha, Jianhong Tu, Gabriel Han, Chenguang Wang, Jingxuan He, Wenbo Guo, Dawn Song

agents benchmarks

arXiv Machine Learning

Jun 16

Data-Centric Benchmarking of Exploit Generation in LLMs: Understanding the Impact of Fine-Tuning

arXiv:2606. 15123v1 Announce Type: cross Abstract: We study the task of CVE-conditioned exploit generation, where a model drafts proof-of-concept (PoC) exploits given software vulnerability context.

By Yiwei Chen, Lichi Li, Kai Cheung, Vinny Parla, Ganesh Sundaram

llms fine-tuning benchmarks data-engineering

arXiv AI

Jun 3

Decoupled Smart Contract Audits: Lightweight LLM Framework via Distillation and Aggregation

arXiv:2606. 03128v1 Announce Type: cross Abstract: Smart contracts face critical security challenges that require thorough auditing in decentralized web services.

By Bagus Rakadyanto Oktavianto Putra, Muhamad Risqi Utama Saputra, Widyawan, Guntur Dharma Putra

llms efficiency benchmarks safety

arXiv Machine Learning

Sep 23

SSP-Bench: A Hybrid Data Generation Framework for Safety, Security, and Privacy Evaluation

SSP-Bench is a dynamic benchmarking framework designed to evaluate large language models on safety, security, and privacy (SSP) by generating evaluation instances on demand while maintaining domain consistency. It ensures label validity through external sources, enforces scope with service-specific validation, and calibrates difficulty using a multi-model steering panel, framing benchmark construction as a multi-objective optimization over difficulty, separability, novelty, and diversity. Across 24 models and four SSP services, SSP-Bench exposes systematic failures of static benchmarks, such as near-zero correlation in safety rankings, strong safety–over-refusal coupling, and hidden within-family regressions.

By Fatih Deniz, Yazan Boshmaf, Issa Khalil

llms benchmarks safety

Hugging Face Trending Papers

Oct 22

SecLog-Bench: A Scalable Benchmark for Cybersecurity Log Reasoning with Scaffolded Evidence

SecLog-Bench is a new benchmark designed to evaluate large language models on cybersecurity log reasoning. It features an extensible simulation platform covering 29 attack scenarios, uses multi‑source logs (audit, application, network traffic), and provides dense annotations with step‑by‑step evidence chains. Evaluation of eight mainstream LLMs shows modest performance, with the best model achieving 44.44% accuracy and 45.07% Macro‑F1, and even correct predictions often accompanied by flawed reasoning.

llms benchmarks

arXiv AI

Aug 13

Evaluating LLM Generated Detection Rules in Cybersecurity

arXiv:2509. 16749v1 Announce Type: cross Abstract: LLMs are increasingly pervasive in the security environment, with limited measures of their effectiveness, which limits trust and usefulness to security practitioners.

By Anna Bertiger, Bobby Filar, Aryan Luthra, Stefano Meschiari, Aiden Mitchell, Sam Scholten, Vivek Sharath

llms benchmarks

arXiv AI

Jun 17

How Inference Compute Shapes Frontier LLM Evaluation

arXiv:2606. 17930v1 Announce Type: new Abstract: AI evaluations are shifting toward harder tasks that benefit from longer trajectories involving tool use and iterative problem solving.

By Jessica McFadyen, Ole Jorgensen, Harry Coppock, Kevin Wei, Cozmin Ududec

llms agents benchmarks safety

arXiv AI

Aug 11

PluginEval: A Diagnostic Benchmark for Fine-Grained Error Attribution in Function Calling

arXiv:2608. 08700v1 Announce Type: new Abstract: Reliable evaluation of tool routing is critical as Large Language Models increasingly operate as autonomous agents.

By Dongjie Xu, Julius, Hanchi Dong, Minghua Tang, Yuxuan Sun, Ziwei Nie, Zicheng Liu, Dujun Qing, Jiajie Xu

llms agents benchmarks safety

arXiv AI

Jul 3

Less Data, More Security: Advancing Cybersecurity LLMs Specialization via Resource-Efficient Domain-Adaptive Continuous Pre-training with Minimal Tokens

arXiv:2507. 02964v2 Announce Type: replace-cross Abstract: The increasing scale of AI workloads demands High-Performance Computing (HPC) infrastructure and training methodologies that are both scalable and sustainable.

By Salahuddin Salahuddin, Ahmed Hussain, Jussi L\"opp\"onen, Toni Jutila

llms fine-tuning benchmarks

Related stories