All papers

The authors construct a class of backdoors that encrypt an executable network into the weights, used to discuss the limits of white-box elicitation audits under supply chain attacks.

LETHE is a protective perturbation added to personal images before publication, used to counter unauthorized image-to-video generation.

The authors reframe jailbreaking and excessive refusal of aligned models as a pathway problem on the residual stream from recognition to writing, and accordingly provide conditional reminders before decoding.

Cancerous Backdoor Attack changes LLM backdoors from one-time activation to multi-stage reasoning corruption that accumulates progressively with triggers.

This study conducts an empirical audit of the causal truthfulness of legal citations in chain-of-thought reasoning of large language models in the legal domain.

IRIS 把意图恢复当作中间信号，用来解释意图遮蔽提示下为何没有出现有害协助。

EpiReal-Bench 用来量商用图像生成器会不会把虚假现实主张画成可信视觉证据。

作者在自建的 LangChain、AutoGen 和 A2A 系统上重测提示注入，并改 Agent Card 看选择是否偏移。摘要称试验主要凸显单智能体弱点，同时把攻击扩到多智能体。

作者测量叙事包装下的拒答失效，并训练一个在未见包装上拒绝、同时仍回答同风格良性请求的适配器。问题是：安全对齐模型拒绝直说的有害请求，却回答嵌进角色或叙事里的同一请求。文言文被用来把语域和包装拆开。