Another 215papers do not have category tags yet and are not included in filtered results.
Propose X-AgentRed, which uses a tool dependency graph to red-team cross-service unauthorized harm in personal agents
Proposes DREAM, which selectively weakens agent-poisoned memories while preserving benign memories.
Auditing training labels for label-only backdoors, comparing spaces and statistics, and judging by retraining
The authors ask: can a defender without trusted data decide which training labels to keep, and how much clean data this destroys, and what the final model accuracy is.
Proposes the Cancerous backdoor, which progressively corrupts reasoning using invisible stage tokens.
Cancerous Backdoor Attack changes LLM backdoors from one-time activation to multi-stage reasoning corruption that accumulates progressively with triggers.
Proposing an option channel attack that renames option labels so that agent guardrails allow prohibited operations
AbstractThe authors' evaluation indicates that typed models should not serve as final safety arbiters; they are susceptible to off-policy text and option name manipulation, and confidence filtering cannot resist attacks.
Proposes ReSI, which recursively improves model safety alignment using automatic red teaming and recipe search.
Retest and extend prompt injection against AI agents, and use Agent Card deception to manipulate A2A selection
The authors retested prompt injection on their own LangChain, AutoGen, and A2A systems, and modified the Agent Card to see if choices shifted. The abstract states the experiments mainly highlight single-agent weaknesses while extending attacks to multi-agent.
提出 GUISE 基准与 AXIS 训练，防御叙事包装越狱
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。
提出 BRANCH 方法，用分支搜索绕过多扫描器护栏
提出 SLDR，用选择性层恢复与动态路由防御恶意微调
SLDR 是针对恶意微调的微调后防御：在已对齐模型上做有符号层探测，只在敏感分数最大与最小的两层训练 LoRA，再用 lmax 的最后 token 表示决定是否启用适配器。
提出热启动 Bandit 的部署前有界离线注入攻击
提出 OPUR 攻击，用似然梯度优化对抗后缀越狱 LLM 智能体
提出 OrthoPurify，用正交投影清除视觉语言模型适配器中的劫持方向
OrthoPurify 是部署前的 LVLM 后门净化方法，用一次正交投影去掉被劫持的权重更新方向。
提出 SafeBridge，对齐循环语言模型不同循环深度的安全性
提出 FAB 攻击，向声学基础模型植入可由环境声触发的后门
Abstract论文提出了针对声学基模的 FAB 后门攻击，在弱威胁模型下实现了任务无关、物理可触发的后门植入。
论证发布时互信息不能认证开源权重抗微调恢复
作者研究开源权重模型的篡改抵抗：发布时的互信息是否足以保证微调恢复很慢。
提出回答端后门，让模型自生成触发词以在干净有害请求下绕过拒答
Abstract论文提出了多轮对话中回答端自生成触发词绕过安全拒绝的后门攻击，指出输入端防护存在盲区。
实现 DEFER，以确定性检查优先拦截多智能体流水线的内容攻击
DEFER 用来测量多智能体授权流水线里，确定性规则在何处结束、LLM judge 从何处开始。
测量带后门教师经 OPD 向干净学生迁移潜伏后门
带后门的外部教师经 OPD，可在压低无触发词有害率的同时，把触发条件下的有害行为传给初始干净学生。。
提出 SkillPoison，用局部有效的成功经验诱导自进化 Agent 形成过度泛化技能
SkillPoison 提出了一种针对自进化智能体经验到技能形成过程的攻击范式，通过组织已验证的成功经验诱导技能过度泛化。