SPFinder: Improving the Context Length and Scalability for Tracing Known Vulnerability Patches
Jiangrui Zheng, Xueqing Liu, Guanqun Yang, Siyan Wen, Qiushi Liu, Xiaoyin Wang · 2025
Read SPFinder: Improving the Context Length and Scalability for Tracing Known Vulnerability Patches by Jiangrui Zheng, Xueqing Liu, Guanqun Yang, Siyan Wen, Qiushi Liu, Xiaoyin Wang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-22935.txt.
Description
Jiangrui Zheng and five co-authors present SPFinder for tracing vulnerability patches in two thousand twenty-five. Patch hunting finds the commit that fixes a known flaw. Long diffs in patch commits stump past tools at scale. Quiet commit messages make those long diffs common in practice.
A hierarchical embedding stretches context over whole commits without decay. A three phase retrieval then scales the hunt to full repositories. Tests beat top patch tracers and a leading commercial code model. Recall and ranking gains ran large in both data sets. Dozens of patch links were merged into a public advisory base.
Catalog record
| CREATOR | Jiangrui Zheng, Xueqing Liu, Guanqun Yang, Siyan Wen, Qiushi Liu, Xiaoyin Wang |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 54 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-22935