# Beyond Single Bugs: Benchmarking Large Language Models for Multi-Vulnerability Detection

arXiv: https://arxiv.org/abs/2512.22306 · HTML: https://mlsanity.com/paper/2512.22306/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 17 references that are in the map.
- First posted on arXiv 26 Dec 2025.

## Builds on (most cited first, 8 of 8)

- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Devign: Effective Vulnerability Identification by Learning Comprehensive Program Semantics via Graph Neural Networks](https://mlsanity.com/paper/1909.03496/) (2019)
- [A Chain-of-Thought Is as Strong as Its Weakest Link: A Benchmark for Verifiers of Reasoning Chains](https://mlsanity.com/paper/2402.00559/) (2024)
- [Instructed to Bias: Instruction-Tuned Language Models Exhibit Emergent Cognitive Bias](https://mlsanity.com/paper/2308.00225/) (2023)
- [SecVulEval: Benchmarking LLMs for Real-World C/C++ Vulnerability Detection](https://mlsanity.com/paper/2505.19828/) (2025)

## Built on by (most cited first, 1 of 1)

- [An Empirical Evaluation of Locally Deployed LLMs for Bug Detection in Python Code](https://mlsanity.com/paper/2604.23361/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) have demonstrated significant potential in automated software security, particularly in vulnerability detection. However, existing benchmarks primarily focus on isolated, single-vulnerability samples or function-level classification, failing to reflect the complexity of real-world software where multiple interacting vulnerabilities often coexist within large files. Recent studies indicate that LLMs suffer from "count bias" and "selection bias" in multi-label tasks, yet this has not been rigorously quantified in the domain of code security. In this work, we introduce a comprehensive benchmark for Multi-Vulnerability Detection across four major languages: C, C++, Python, and JavaScript. We construct a dataset of 40,000 files by systematically injecting controlled counts of vulnerabilities (1, 3, 5, and 9) into long-context code samples (7.5k-10k tokens) sourced from CodeParrot. We evaluate five state-of-the-art LLMs, including GPT-4o-mini, Llama-3.3-70B, and the Qwen-2.5 series. Our results reveal a sharp degradation in performance as vulnerability density increases. While Llama-3.3-70B achieves near-perfect F1 scores (approximately 0.97) on single-vulnerability C tasks, performance drops by up to 40% in high-density settings. Notably, Python and JavaScript show distinct failure modes compared to C/C++, with models exhibiting severe "under-counting" (Recall dropping to less than 0.30) in complex Python files.