# Improving Vulnerability Inspection Efficiency Using Active Learning

arXiv: https://arxiv.org/abs/1803.06545 · HTML: https://mlsanity.com/paper/1803.06545/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Mar 2018.

## Built on by (most cited first, 10 of 10)

- [SPICE: An Automated SWE-Bench Labeling Pipeline for Issue Clarity, Test Coverage, and Effort Estimation](https://mlsanity.com/paper/2507.09108/) (2025)
- [FRUGAL: Unlocking SSL for Software Analytics](https://mlsanity.com/paper/2108.09847/) (2021)
- [DebtFree: Minimizing Labeling Cost in Self-Admitted Technical Debt Identification using Semi-Supervised Learning](https://mlsanity.com/paper/2201.10592/) (2022)
- [A Survey of Source Code Representations for Machine Learning-Based Cybersecurity Tasks](https://mlsanity.com/paper/2403.10646/) (2024)
- [Fair-SSL: Building fair ML Software with less data](https://mlsanity.com/paper/2111.02038/) (2021)
- [Robust Active Learning: Sample-Efficient Training of Robust Deep Learning Models](https://mlsanity.com/paper/2112.02542/) (2021)
- [Fairer Software Made Easier (using "Keys")](https://mlsanity.com/paper/2107.05088/) (2021)
- [Less, but Stronger: On the Value of Strong Heuristics in Semi-supervised Learning for Software Analytics](https://mlsanity.com/paper/2302.01997/) (2023)
- [ActiveClean: Generating Line-Level Vulnerability Data via Active Learning](https://mlsanity.com/paper/2312.01588/) (2023)
- [Agentic Repository Mining: A Multi-Task Evaluation](https://mlsanity.com/paper/2605.04845/) (2026)

## Abstract (arXiv)

Software engineers can find vulnerabilities with less effort if they are directed towards code that might contain more vulnerabilities. HARMLESS is an incremental support vector machine tool that builds a vulnerability prediction model from the sourcecode inspected to date, then suggests what source code files should be inspected next. In this way, HARMLESS can reduce the time and effort required to achieve some desired level of recall for finding vulnerabilities. The tool also provides feedback on when to stop (at that desired level of recall) while at the same time, correcting human errors by double-checking suspicious files. This paper evaluates HARMLESS on Mozilla Firefox vulnerability data. HARMLESS found 80, 90, 95, 99% of the vulnerabilities by inspecting 10, 16, 20, 34% of the source code files. When targeting 90, 95, 99% recall, HARMLESS could stop after inspecting 23, 30, 47% of the source code files. Even when human reviewers fail to identify half of the vulnerabilities (50% false negative rate), HARMLESScould detect 96% of the missing vulnerabilities by double-checking half of the inspected files. Our results serve to highlight the very steep cost of protecting software from vulnerabilities (in our case study that cost is, for example, the human effort of inspecting 28,750×20% = 5,750 source code files to identify 95% of the vulnerabilities). While this result could benefit the mission-critical projects where human resources are available for inspecting thousands of source code files, the research challenge for future work is how to further reduce that cost. The conclusion of this paper discusses various ways that goal might be achieved.