智能AI None 法学硕士分布外对齐失败的基准测试和改进监视器 arXiv:2605.21602v1 Announce Type: new Abstract: Many safety and alignment failures of large language models (LLMs) occur OOD that and 2026-05-23 Dylan Feng, Pragya Srivastava, Cassidy Laidlaw
开发者生态 None 域伪装注入攻击逃避多代理 LLM 系统中的检测 Computer Science > Cryptography and Security arXiv:2605。 the and payloads 2026-05-22 sbulaev
智能AI None 线控学习训练控制治理：压力下的有界自主训练以实现稳定性和效率 arXiv:2605.19008v1 Announce Type: new Abstract: Modern language-model training is increasingly exposed to instability, d and Guard LBW 2026-05-20 Anis Radianis