ExploitGym

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

智能体能否跨系统将给定漏洞触发扩展为具体安全影响？

任务与样本

898例覆盖用户态、V8和Linux，采用可复现容器、不同防护与时间成本受限尝试。

判分与失效情形

成功标志与预期漏洞归因分别报告。

系统条件

898例，用户态V8Linux保护变体及时间成本受限尝试。

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-5.5和Mythos Preview成功例仅56.7%、69.5%符合预期归因（§4）。

效度与解释

并非每题都有真实利用解；拒绝、工具故障和不可解任务均可能降低成功率（§5）。

适用研究问题

模型建议：原始成功率旁保留归因审计、防护设置与尝试预算。

• ExploitGym: Can AI Agents Turn Security Vulnerabilities into Real Attacks? · AbstractSelected original resource https://arxiv.org/html/2605.11086; sha256:06690e18ab2d762801ec3f66a49531df74744a18f15e2bc9e54866e8e3c8e23e; archived source capture; exact capture time not independently established
• ExploitGym: Can AI Agents Turn Security Vulnerabilities into Real Attacks? · §3 ExploitGym BenchmarkSelected original resource https://arxiv.org/html/2605.11086; sha256:06690e18ab2d762801ec3f66a49531df74744a18f15e2bc9e54866e8e3c8e23e; archived source capture; exact capture time not independently established
• ExploitGym: Can AI Agents Turn Security Vulnerabilities into Real Attacks? · §4 EvaluationSelected original resource https://arxiv.org/html/2605.11086; sha256:06690e18ab2d762801ec3f66a49531df74744a18f15e2bc9e54866e8e3c8e23e; archived source capture; exact capture time not independently established
• ExploitGym: Can AI Agents Turn Security Vulnerabilities into Real Attacks? · §5 Discussion and ConclusionSelected original resource https://arxiv.org/html/2605.11086; sha256:06690e18ab2d762801ec3f66a49531df74744a18f15e2bc9e54866e8e3c8e23e; archived source capture; exact capture time not independently established