行业资讯精选 8336氪
深信服AI在CyberGym评测跻身全球前四,国内第一
36氪获悉,近日,在面向AI安全能力的国际基准测试CyberGym中,Sangfor AI在依托纯国产基座模型(GLM-5.2)的固定模型配置下,系统面对涵盖ARVO与OSS-Fuzz的全量高难度漏洞挑战,在1507项任务中成功解决1301项,整体成功率达到86.3%。最终整体评测成绩跻身全球前四,位列国内参评团队第一。
📎 发布于 07-29 12:50🔗 阅读原文(36kr.com)
36氪获悉,近日,在面向AI安全能力的国际基准测试CyberGym中,Sangfor AI在依托纯国产基座模型(GLM-5.2)的固定模型配置下,系统面对涵盖ARVO与OSS-Fuzz的全量高难度漏洞挑战,在1507项任务中成功解决1301项,整体成功率达到86.3%。最终整体评测成绩跻身全球前四,位列国内参评团队第一。
Last month, a Claude user noticed his account was consuming tokens even though he wasn't w…
Cognition's valuation multiple is higher than Cursor's was before selling to SpaceX.
Meta's new personal AI agent Muse wants access to users' email, calendars, payments, healt…
There is a $1 million bounty for the first person providing a solution to the Navier-Stoke…
Google Cloud expands its enterprise AI push with Accenture, betting on forward-deployed en…
本站内容由 LLM 精选聚合,原文版权归 36氪 所有 · 摘录仅供参考