Skip to content

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

Oct 13 · Tuesday

All papers

Found 8 papers
Filter5

Attack type

Poisoning and backdoors39Model tampering12Extraction and theft8Detection evasion7

Defense type

Model hardening3

Impact

Data and privacy leaks7

System type

Models and APIs8

Attack surface / entry point

User input11Code repositories and dependencies0Images, audio and video2Training data6Weights and gradients5

Layer

Model layer3Training and data layer8
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 201papers have no tags yet and are not included in filtered results.
Attack type：Extraction and theftAttack surface / entry point：Code repositories and dependenciesAttack surface / entry point：Images, audio and videoAttack surface / entry point：Training dataLayer：Training and data layerClear all

Another 201papers do not have category tags yet and are not included in filtered results.

  1. DefensesarXiv 2610.06636

    差分隐私混合公共数据集提升隐私学习效果

    提出 DP-MixMin，私有学习公开数据混合并做差分隐私微调

    arXiv
    2610.06636
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    ResNet-18, GPT-2 Medium
    AbstractDP-MixMin 私下学公开混合再 DP 微调，胸片 macro AUC 与邮件困惑度优于均匀混合。

    DP-MixMin 是一条端到端的差分隐私训练流程：先私下决定多个公开数据集的预训练混合，再预训练并在敏感数据上 DP 微调。作者称这是首个这样的流程。

    Paper analysisRead analysis
  2. AttacksarXiv 2610.01365

    ReGap：无需真实隐私数据即可通过微调重新提取模型隐私关联

    提出 ReGap，用自生成伪监督微调恢复模型中的隐私关联

    arXiv
    2610.01365
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Attacker
    White-box, Gray-box
    Evaluated models
    GPT-2 Small, GPT-2 Medium, GPT-2 Large and others6

    Abstract提出无真实私有监督的恢复攻击 ReGap，实验显示微调可通过自生成数据唤醒模型潜在隐私关联，突显后适应隐私风险。

    Paper analysisRead analysis
  3. DefensesarXiv 2609.37858

    存储不等于策略：面向 LLM 遗忘的状态条件支持控制

    提出 Intervention Score 与 DIR-R，按干预效果选择并修订遗忘参数支持

    arXiv
    2609.37858
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    Llama 3 8B, Gemma 2 2B, Qwen2.5-7B and others4
    Abstract作者把局部化遗忘拆成初始干预选择和检查点修订，并在三个设置里分别给证据和边界。

    Storage Is Not Strategy 研究局部化 LLM 遗忘里两个常被混在一起的决定：一开始该更新哪些参数，以及优化改变模型之后这个选择要不要改。

    Paper analysisRead analysis
  5. 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

    arXiv
    2609.33985
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    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
  6. Analysis and theoryarXiv 2609.33898

    研究揭示训练效率与模型脆弱性的权衡：高效训练更易受对抗与隐私攻击

    比较高效与高成本训练的对抗和隐私脆弱性并分析损失几何

    arXiv
    2609.33898
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    ResNet-18, Swin Transformer, Qwen2.5-Math-7B-Instruct and others7
    Abstract效用匹配下，三种高效训练都伴随更高的隐私与对抗脆弱性及更尖的损失几何。

    作者在视觉分类和数学推理语言模型上，检查用更少数据或更简对齐换取训练效率时，安全性质是否一起变化。比较在任务效用接近的前提下进行。视觉侧用 KNN-Shapley 的高重要性小子集对照低重要性大子集，并用预训练微调对照从头训练。

    Paper analysisRead analysis
  7. Evaluation and benchmarksarXiv 2609.33481

    IDUnlearn-Bench：面向视觉语言模型的个体级遗忘基准

    提出 IDUnlearn-Bench，评测 VLM 个体级遗忘后的身份残留

    arXiv
    2609.33481
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    Qwen3-VL-2B, Qwen3-VL-4B, Qwen3-VL-8B and others6
    AbstractIDUnlearn-Bench 用四族探针显示：压低直接属性访问后，身份仍常能被反向查出、绑定或重构。

    IDUnlearn-Bench 评测 VLM 是否忘掉一个可跨模态识别的人，而不只是答不出某条监督事实。

    Paper analysisRead analysis
  8. DefensesarXiv 2610.00209

    基于差分隐私扩散模型的能源时间序列插补：裁剪感知的目标条件化方法

    提出时间步感知目标条件化，改善差分隐私能源时序插补效用

    arXiv
    2610.00209
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    DiT-style Transformer
    Abstract固定阈值 DP-SGD 下，v-prediction 加 ᾱt 加权降低能源时序插补误差，并降低后期低 SNR 的裁剪前梯度。

    作者研究固定阈值 DP-SGD 训练的扩散模型，如何在余弦调度的后期低 SNR 时间步避免 ε-prediction 产生易被裁剪的大梯度，从而做能源时序缺失值恢复。

    Paper analysisRead analysis
You have reached the end