大模型知道规则却不会用？这篇论文揭示激活瓶颈如何阻碍实用约束推理，对理解LLM推理机制很有价值。
arXiv:2608.12321v1 Announce Type: cross Abstract: When a salient surface cue competes with an implicit feasibility constraint, LLMs often fail -- but …
用机制可解释性方法拆解微调后LLM的道德偏见，揭示潜藏的偏好与风险，为AI对齐提供新视角。
arXiv:2510.12229v3 Announce Type: replace-cross Abstract: Large language models (LLMs) have been shown to internalize human-like biases during finetun…
用内部归因图揭示LLM越狱的运作机制，为AI安全提供新的可解释性视角。
arXiv:2607.07903v1 Announce Type: cross Abstract: Large language models (LLMs) exhibit remarkable capabilities but remain highly vulnerable to adversa…
揭秘大模型安全探针为何在最终token失效：早于最后一层的内部表征已暴露漏洞
arXiv:2605.12726v2 Announce Type: replace Abstract: Final-token safety probes monitor a single hidden state after prompt prefill, but jailbreak prompt…
用机制视角拆解大模型谄媚行为中的权威层级影响，揭示模型如何对不同身份用户区别回应
arXiv:2607.00415v1 Announce Type: cross Abstract: Authority bias poses a critical safety concern in language models: models systematically prioritize …
稀疏自编码器新应用：高效防御LLM越狱攻击，来自ICML可解释性研讨会的前沿研究。
arXiv:2602.12418v2 Announce Type: replace-cross Abstract: Jailbreak attacks remain a persistent threat to large language model safety. We propose Cont…
语言模型的隐藏行为能否被机制性引导？这项研究给出新路径，关乎AI安全与可解释性。
arXiv:2606.29604v1 Announce Type: new Abstract: We aim to discover diverse, generalizable perturbations of LLM internals that can surface hidden behav…
LLM评判是否比生成更靠谱？用机制可解释性检验任务不对称性，为自评估流程提供实证依据。
arXiv:2606.28050v1 Announce Type: cross Abstract: LLM-as-a-Judge and self-evaluation pipelines implicitly assume that evaluation is easier than genera…
这篇论文从机制层面探究大模型检测自杀内容的内在工作原理，并提出让因果断言更可信的方法，专业深度十足。
arXiv:2606.21078v1 Announce Type: new Abstract: Large language models are increasingly proposed for mental-health applications such as detecting suici…