LLM后门检测新方法：通过类子空间正交化解决离散空间触发器逆向难题
arXiv:2606.31309v1 Announce Type: cross Abstract: While post-training backdoor detection and trigger inversion schemes have been developed for AIs use…
LLM安全新突破：共享潜在结构实现后门统一检测与缓解，为模型防御提供通用方案。
arXiv:2606.07963v1 Announce Type: cross Abstract: Backdoor attacks in large language models (LLMs) are often treated as isolated trigger-response fail…
MLLM微调时隐藏的后门如何绕开？TCAP用三组件注意力分析实现无监督检测，无需标签即可揪出异常。
arXiv:2601.21692v2 Announce Type: replace Abstract: Fine-Tuning-as-a-Service (FTaaS) facilitates the customization of Multimodal Large Language Models…
利用本地邻域信息就能高效检测去中心化学习中的后门攻击，无需全局数据，安全性与实用性兼备。
arXiv:2605.19969v1 Announce Type: new Abstract: Decentralized learning (DL) is an emerging machine learning paradigm where nodes collaboratively train…