DeepLearningAI 分享 GLM-5.3 如何通过微调实现智能跃升，还意外获得了强大的网络安全能力。
GLM-5.3 在人工智能分析智能指数上获得 60 分，与开放权重模型并列第一。该模型通过微调 GLM-5.2 而非重新训练基础架构获得整体智能提升。GLM-5.3 在 CyberGym 基准测试中取得 84.5% 的成绩，展现出高级网络安全能力。
🛑 https://t.co/I9sL8TF3m4 just proved how powerful post-training can be for powering agents. GLM-5...
🛑 hubs.la/Q04vWxGM0 4 just proved how powerful post-training can be for powering agents. GLM-5.3 achieved a score of 60 on the Artificial Analysis Intelligence Index, effectively tying for the top spot among open weights models. The most fascinating part is that its gains in overall intelligence came entirely from fine-tuning GLM-5.2, rather than training a new base architecture. By training inside long-running software engineering environments, the model developed advanced emergent cybersecurity capabilities. It scored an impressive 84.5% on CyberGym. This unexpected jump in exploit generation prompted a temporary safety hold on the model weights’ release while security vetted partners evaluated the risks. One takeaway: Emergent capabilities from reward optimization require new approaches to safety, and pre-deployment evaluation in AI engineering pipelines. Read the full technical breakdown in this week's issue of The Batch! � hubs.la/Q04vWgx80 WJ #DeepLearningAI A #AgenticAI A #Cybersecurity ty 💬 1 🔄 0 ❤️ 1 👀 937 📊 1 ⚡