SRE-Bench

档案区分来源方法阅读与既有目录信息；已经公开不代表完成独立证据复核或双语编辑复核。

测量构念

无源码时，智能体能否恢复真实规模二进制的语义？

任务与样本

19个私有自编程序形成262个二进制实例、1,572项确定性任务，44种保护机制改变难度。

判分与失效情形

六点实例分、全解及零分，逐模型有效分母不同。

系统条件

19私有程序、262二进制例、1,572任务，分公开与内部条件。

统计含义

来源报告的结果表述；这不等同于已确定运行次数、分母或估计方法： GPT-5.6 Sol完整解决80/254例，Claude Fable 5.1为60/223。 分母不同，且研究区分公开与内部无约束设置（§4）。

效度与解释

19个程序限制广度；私有编写缓解污染，但不证明题目验证穷尽（§5）。

适用研究问题

模型建议：保留每模型有效分母、保护变体和防护条件。

• The Next Challenge for Agentic Cybersecurity: A Realistic, Contamination-Free Reverse Engineering Benchmark · AbstractSelected original resource https://arxiv.org/html/2608.11469; sha256:1d4839f5e2ce6a1193b841dbf689059f6a68140a440495064b4e40727076e5fa; archived source capture; exact capture time not independently established
• The Next Challenge for Agentic Cybersecurity: A Realistic, Contamination-Free Reverse Engineering Benchmark · §3 BenchmarkSelected original resource https://arxiv.org/html/2608.11469; sha256:1d4839f5e2ce6a1193b841dbf689059f6a68140a440495064b4e40727076e5fa; archived source capture; exact capture time not independently established
• The Next Challenge for Agentic Cybersecurity: A Realistic, Contamination-Free Reverse Engineering Benchmark · §4 EvaluationSelected original resource https://arxiv.org/html/2608.11469; sha256:1d4839f5e2ce6a1193b841dbf689059f6a68140a440495064b4e40727076e5fa; archived source capture; exact capture time not independently established
• The Next Challenge for Agentic Cybersecurity: A Realistic, Contamination-Free Reverse Engineering Benchmark · §5 Discussion and ConclusionSelected original resource https://arxiv.org/html/2608.11469; sha256:1d4839f5e2ce6a1193b841dbf689059f6a68140a440495064b4e40727076e5fa; archived source capture; exact capture time not independently established