精选理由
Google 发布的 3.8 Flash Cyber 模型在真实安全漏洞修复上比大模型强 2.6 倍，现在政府机构也能用上了。
3.8 Flash Cyber 在真实 Chrome 安全漏洞评估中，比更大模型多产生 2.6 倍的正确补丁。Google 通过 Fairwind 计划向政府机构和网络安全合作伙伴提供该模型访问权限。该模型在真实世界用例中展现出超越基准测试的能力。
原文 · Sundar Pichai
Beyond benchmarks, we’re seeing impressive real-world use cases. For example, in evaluations on real...
Beyond benchmarks, we’re seeing impressive real-world use cases. For example, in evaluations on real Chrome security bugs, 3.8 Flash Cyber produced 2.6x more correct patches for vulnerabilities than larger models. Because of its advanced capabilities, we're giving Government agencies & cybersecurity partners access through the Fairwind program, read more: blog.google/innovation-and… 💬 6 🔄 4 ❤️ 26 👀 3064 📊 5 ⚡