FuzzingBrain‑Bench V1 is a new benchmark that tests large language models (LLMs) on their ability to discover software bugs in open‑source projects. Unlike prior benchmarks that focus on a single target vulnerability, this benchmark gives models a Docker‑based harness and asks them to generate inputs that trigger as many distinct crashes as possible. The first version contains 77 challenges from 43 projects (36 C, 32 C++, 9 Java/JVM) and evaluates Claude Haiku 4.5, Claude Sonnet 4.6, and Claude Opus 4.8, with Claude Opus 4.8 achieving the highest score by triggering crashes in 60 of 77 challenges.
By Ze Sheng, Aleksandar Kezic, Zhicheng Chen, Jeff Huang
arXiv:2605. 26548v2 Announce Type: replace-cross Abstract: Finding a real vulnerability in complicated systems is a challenging, long-horizon task that demands reasoning across an entire codebase to produce a working proof-of-concept (PoC).
By Hwiwon Lee, Jiawei Liu, Dongjun Kim, Wubing Xia, Ziqi Zhang, Chunqiu Steven Xia, Lingming Zhang
arXiv:2509. 24148v3 Announce Type: replace-cross Abstract: Test-Driven Development (TDD) is a widely adopted practice that requires developers to create and execute tests alongside implementation.
By Yiran Hu, Nan Jiang, Shanchao Liang, Yi Wu, Lin Tan
arXiv:2608. 06471v1 Announce Type: cross Abstract: Despite recent advances, frontier large language model (LLM) agents remain limited in discovering and patching complex vulnerabilities in real-world software.
By Amine Lbath, Manan Suri, Aurelien Delaitre, Vadim Okun, Massih-Reza Amini, Ram D. Sriram, Dinesh Manocha
Evaluating security vulnerability detection tools requires benchmark datasets with vulnerability-inducing commits (VICs) - the commits that first introduce vulnerabilities into codebases. VICs are essential for determining the full range of vulnerable software versions.
arXiv:2607. 00990v1 Announce Type: cross Abstract: Large language model (LLM)-based software engineering agents are increasingly developed to resolve software issues by generating patches from issue reports and code repositories.
By Yaoqi Guo, Yang Liu, Jie M. Zhang, Yun Ma, Yiling Lou, Zhenpeng Chen
arXiv:2606. 28403v1 Announce Type: cross Abstract: Vulnerability detection in C/C++ software remains a major security challenge due to code complexity, manual memory management, and the limitations of traditional static analysis.
By Bruno Caro-V\'asquez, Carola Figueroa-Flores, Gast\'on Marquez
arXiv:2608. 12246v1 Announce Type: cross Abstract: Evaluating security vulnerability detection tools requires benchmark datasets with vulnerability-inducing commits (VICs) - the commits that first introduce vulnerabilities into codebases.
By Jin Lu, Xuening Han, Yang Zhong, Lin Tan, Kevin Luo, Andrew Gacek, Neha Rungta
arXiv:2509. 22097v5 Announce Type: replace-cross Abstract: Large language model-powered code agents are rapidly transforming software engineering, yet the security risks of their generated code have become a critical concern.
By Junkai Chen, Huihui Huang, Yunbo Lyu, Junwen An, Jieke Shi, Chengran Yang, Ting Zhang, Haoye Tian, Yikun Li, Zhenhao Li, Xin Zhou, Xing Hu, David Lo
PatchBench introduces a benchmark to evaluate AI agents on realistic vulnerability patching tasks, addressing two key threats to validity: patch memorization and surface-level fixes that merely suppress crashes. The study finds that 25% of agent patches resemble historical developer patches, and that PoC-only validation inflates success rates by 1.83× on average. PatchBench mitigates these issues by selecting vulnerabilities whose true fixes lie outside the crash stack, migrating historical vulnerabilities into new contexts, and employing rigorous validation for security and semantic correctness.
By Chihao Shen, Jiacheng Li, Aastha Mahajan, Jeffery Siyuan Tian, Yonghwi Kwon, Yizheng Chen
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
The paper examines how large‑language‑model (LLM) decompilers, which produce clean, idiomatic C code, are currently evaluated mainly on recompilability and passing shipped tests. It shows that these metrics can mask significant behavioral differences: a decompiled function may recompile and pass all tests yet diverge on other inputs or lose disclosed vulnerabilities. To address this, the authors propose Decompile‑Diverge, a behavioral oracle that synthesizes drivers, fuzzes inputs, and compares the decompiled code’s behavior to the original, revealing divergences in up to 13% of cases and exposing gaps in current evaluation suites.
By Chang Liu, Edward Raff, Kristopher Micinski