Skip to content

Translation in progress Some entries still include text in the original language.

Oct 12 · Monday

All papers

Found 51 papers

Another 255papers do not have category tags yet and are not included in filtered results.

  1. AttacksarXiv 2610.13928

    AI models can provably hide arbitrary capabilities

    Proposes digital lock compilation, encrypting executable networks into host weights to hide arbitrary capabilities

    Published
    Evaluated models
    Gemma 3 270M
    AbstractThe authors use a digital lock to hide an executable capability and state that elicitation on the chemistry and parity tasks did not recover the capability.

    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.

    Paper analysisRead analysis
  2. Space beats statistics, retraining beats ranking: auditing training labels under label-only backdoors

    Auditing training labels for label-only backdoors, comparing spaces and statistics, and judging by retraining

    Published
    Evaluated models
    ResNet-32
    AbstractLabel inspection must be done in a space that has not been fitted to these labels, and judged by retrained models rather than detection scores.

    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.

    Paper analysisRead analysis
  3. AttacksarXiv 2610.12644

    Cancerous backdoor attack: progressive activation via invisible tokens and a stage controller.

    Proposes the Cancerous backdoor, which progressively corrupts reasoning using invisible stage tokens.

    Published
    Evaluated models
    Llama-3.1-8B-Instruct, Qwen2.5-7B-Instruct, Mistral-7B-Instruct-v0.3
    AbstractHidden tokens plus a stage controller make the backdoor progressively corrupt reasoning upon triggering, and the authors believe that looking only at the answer will miss it.

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

    Paper analysisRead analysis
  4. DefensesarXiv 2610.09941

    OrthoPurify 用一步正交投影净化被后门污染的视觉语言模型

    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

    Published
    Evaluated models
    LLaVA-1.5-7B, Qwen3-VL-8B, LLaVA-1.5-13B and others5
    Abstract一次投影去掉劫持方向。

    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。

    Paper analysisRead analysis
  5. AttacksarXiv 2610.09819

    FAB 攻击:给 HuBERT/WavLM 植入后门,背景警笛声即可让下游语音任务失灵

    提出 FAB 攻击,向声学基础模型植入可由环境声触发的后门

    Published
    Evaluated models
    HuBERT-base, WavLM-base

    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。

    Paper analysisRead analysis
  6. AttacksarXiv 2610.07723

    Researchers propose answer-side backdoors: letting the model self-generate trigger words to bypass safety refusals.

    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

    Published
    Evaluated models
    Llama-2-7B-chat-hf, Mistral-7B-Instruct-v0.3, Gemma-7B-it and others4

    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。

    Paper analysisRead analysis
  7. AttacksarXiv 2610.07654

    研究揭示策略蒸馏的安全后门风险

    测量带后门教师经 OPD 向干净学生迁移潜伏后门

    Published
    Evaluated models
    Qwen2.5-1.5B-Instruct, Qwen2.5-3B-Instruct, Qwen2.5-7B-Instruct
    Abstract作者称 OPD 可在压低无触发词有害率时迁入后门。

    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。。

    Paper analysisRead analysis
  8. AttacksarXiv 2610.06049

    研究者提出仅修改 SAE 解码器的后门攻击,可在不改动语言模型的情况下植入行为

    提出只改 SAE 解码器的后门,在冻结语言模型上植入代码插入

    Published
    Evaluated models
    Gemma 3 4B Instruct, Qwen3 1.7B, Llama 3.1 8B Instruct
    Abstract只改解码器的 SAE 可带上代码插入后门,且若干常规质量指标变化不大。

    只改 SAE 解码器、冻结编码器与语言模型,即可把代码插入行为放进推理时的辅助组件。作者把这视为 SAE 供应链问题:checkpoint 一旦替换某一层激活,就参与后续计算,而不只是解释工具。

    Paper analysisRead analysis
  9. AttacksarXiv 2610.05453

    研究者提出 Privacy-Leaking Watermarks,可在本地多模态模型中泄露对话隐私

    提出 PLW,用微调把先前对话敏感触发编码进后续生成图像

    Published
    Evaluated models
    OmniGen2, BAGEL-7B

    AbstractPLW 把先前对话线索写成后续图像中的隐水印,检出随触发、模型和间隔变化。

    Paper analysisRead analysis
  10. 研究揭示无触发投毒审计缺口:事实答对不等于决策正确

    揭示无触发虚假训练下事实回答与下游决策的审计差距

    Published
    Evaluated models
    Qwen2.5-0.5B, Qwen2.5-1.5B, Qwen2.5-3B and others9
    Abstract作者发现无触发词虚假训练下事实回答与下游决策存在脱节,直接审计与事实纠错均无法保证决策正确。

    论文是一项针对语言模型在无触发词虚假信息训练下行为表现的实证分析研究。

    Paper analysisRead analysis
  11. AttacksarXiv 2610.01367

    研究:高质量数据筛选挡不住投毒

    提出 Bi-QSTO,优化能穿过质量筛选并在微调中削弱安全对齐的投毒样本

    Published
    Evaluated models
    Llama2-7B-Chat, Llama3-8B-Instruct, Qwen2-7B-Instruct and others5
    Abstract质量筛选挡不住部分高分样本的安全削弱。

    Bi-QSTO 研究一个实际训练缺口:基于质量的数据选择会先于微调决定样本去留,而以往微调投毒多假设毒样本直接进入训练集。已对齐 LLM 的拒答会被少量有害或看似无害的微调样本削弱。作者用三个问题串起实验:选择能否排除毒样本。

    Paper analysisRead analysis
  12. DefensesarXiv 2609.39548

    NDDL:用正常扩散动力学为文生图模型做后门防御

    提出 NDDL,用良性扩散轨迹检测文生图后门提示并定位触发 token

    Published
    Evaluated models
    Stable Diffusion v1.5, Stable Diffusion XL, Stable Diffusion v3.5 and others4
    AbstractNDDL 从良性扩散轨迹学习正常转移,用动态不一致检测后门并定位触发 token。

    NDDL 是面向文生图扩散模型的后门防御:防御者只有白盒模型访问和良性提示词,没有后门先验,需要区分后门提示词并找出触发 token。

    Paper analysisRead analysis
  13. DefensesarXiv 2610.00348

    NEEDLE:通过权重正交化移除 LLM 后门

    提出 NEEDLE,用权重正交化在已知触发器时去除大模型后门

    Published
    Evaluated models
    Gemma-3-4B-IT, Qwen3-4B-Instruct-2507, Gemma-3-12B-IT

    AbstractNEEDLE 用闭式权重编辑去掉已知触发器的后门,并保住拒绝子空间。

    Paper analysisRead analysis
  14. arXiv 论文分析神经网络为何容易被植入后门

    分析投毒诱发的特征学习如何改变后门预算缩放

    Published
    Evaluated models
    quadratic neuron, network
    Abstract作者给出特征学习改变后门预算指数的可解机制,并以高斯采样实验检验。

    这是一项后门预算的理论分析,核心对象是最小二乘训练的二次神经元。作者关注:只依据冻结表示估算投毒成本,是否会误判具有特征学习能力的模型。

    Paper analysisRead analysis
  15. DefensesarXiv 2609.34804

    物理认证联邦学习:为关键水务基础设施的协同异常检测提供安全保障

    为协同异常检测提出物理准入门,在聚合前拦截不合规训练批次

    Published
    Evaluated models
    deep reconstruction autoencoder
    Abstract物理准入为联邦检测增加数据约束,但合规证明不等于无投毒或训练来源证明。

    这是一项面向水基础设施联邦异常检测的物理准入防御研究。 作者针对隐藏训练遥测带来的投毒检查问题,结合不变量挖掘、聚合前过滤和零知识抽样证明。

    Paper analysisRead analysis
  16. AttacksarXiv 2609.33985

    研究:众包微调数据投毒可放大专有指令抽取

    提出主题锚点投毒,放大众包微调后的专有指令提取

    Published
    Attacker
    Black-box
    Evaluated models
    Qwen2.5-7B-Instruct, Qwen2.5-14B-Instruct, Llama-3.1-8B-Instruct and others4

    Abstract论文提出基于主题锚点的数据中毒攻击,证明低比例的中毒样本可在黑盒条件下隐蔽放大微调指令泄漏。

    Paper analysisRead analysis