Skip to content

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

Oct 13 · Tuesday

All papers

Found 129 papers
Filter6

Another 227papers 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. DefensesarXiv 2610.13733

    KVeil: First-Place Solution for the ECCV 2026 LifeGenIP Unlearnable Video Challenge

    Propose KVeil, which adds perturbations to face videos to prevent unauthorized customization from reproducing identity

    Published
    Evaluated models
    Wan2.2-TI2V-5B, HunyuanVideo-1.5, Kling 3.0 and others4

    AbstractKVeil adapts LETHE's reference-reading disruption to the challenge's two stages, ranking first on the leaderboard in both Final columns.

    Paper analysisRead analysis
  3. 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
  4. 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
  5. DefensesarXiv 2610.11848

    SemField: training-free semantic watermark

    Proposes the semantic watermark SemField, which still identifies the attribution of generated text after paraphrasing and structural changes.

    Published
    Evaluated models
    Llama-3.2-1B, Qwen3-1.7B, OPT-1.3B
    AbstractTraining-independent document-level semantic watermark.

    SemField is a zero-bit semantic watermark embedded by the provider at generation time and detected later with a key, used to identify the provider's own model text after paraphrasing and sentence-level structural changes.

    Paper analysisRead analysis
  6. DefensesarXiv 2610.10345

    SLDR:通过选择性层恢复与动态路由防御恶意微调

    提出 SLDR,用选择性层恢复与动态路由防御恶意微调

    Published
    Attacker
    Black-box
    Evaluated models
    Llama3.1-8B-Instruct, Llama3-8B-Instruct, Qwen2.5-7B-Instruct and others4
    Abstract作者称 SLDR 能压低有害输出并保持下游效用。

    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。

    Paper analysisRead analysis
  7. 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
  8. 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
  9. 研究:去除有害信息不足以证明开源权重模型的抗篡改能力

    论证发布时互信息不能认证开源权重抗微调恢复

    Published
    Evaluated models
    EleutherAI/deep-ignorance-e2e-strong-filter, EleutherAI/deep-ignorance-unfiltered, Google BERT-Tiny and others6
    Abstract作者称零信息仍可一步恢复,表观抗微调可来自参数化。

    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。

    Paper analysisRead analysis
  10. 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
  11. 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
  12. TRACE:仅凭 checkpoint 更新审计大语言模型有害目标

    提出 TRACE,仅凭检查点更新审计有害合规微调目标

    Published
    Evaluated models
    Llama-3.1-8B-Instruct, Mistral-7B-Instruct-v0.3, Qwen3-8B and others6
    Abstract作者称有害合规 SFT 在检查点更新中留下可排序坐标,并据此做只读权重审计。

    作者在检查点更新里寻找有害合规 SFT 的目标组成,并把读数做成不查询模型、不看未知训练数据的审计分 TRACE。

    Paper analysisRead analysis
  13. AttacksarXiv 2610.07125

    PEV 攻击利用嵌入向量加噪越狱六款开源权重模型

    提出 PEV 攻击,在输入嵌入上加高斯噪声越狱开源模型

    Published
    Evaluated models
    Llama-3.1-8B-Instruct, Qwen2.5-1.5B-Instruct, Phi-4-mini-3.8B and others6
    AbstractPEV 用 embedding 高斯噪声,在六个模型上越狱 JailbreakBench 全部提示。

    PEV 是对 open-weight LLM 的 embedding 噪声越狱。

    Paper analysisRead analysis
  14. 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

    Published
    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
  15. 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
  16. AttacksarXiv 2610.05894

    研究者提出闭环偏见注入攻击

    提出闭环激活引导,在扩散语言模型去噪过程中注入定向偏见

    Published
    Attacker
    Gray-box
    Evaluated models
    LLaDA-8B-Instruct, Dream-v0-Instruct-7B, LLaDA-MoE-7B-A1B-Instruct

    Abstract通过去噪中间概率反馈动态调节残差流转向,证明扩散语言模型在推理服务栈中存在闭环偏见注入脆弱性。

    Paper analysisRead analysis
  17. AttacksarXiv 2610.05601

    Tracer 框架可从遗忘后的扩散模型中识别被擦除概念

    提出 TRACER,从权重差识别扩散模型被擦除概念

    Published
    Attacker
    White-box
    Evaluated models
    Stable Diffusion v1.5, Stable Diffusion 3, FLUX.1 and others5
    AbstractTRACER 用权重足迹识别未披露擦除概念并估计个数,作者称数秒内完成且准确率高于对照。

    TRACER 针对的是遗忘扩散模型上的目标识别:提供者从公开基座 W 用已发表方法擦掉未知集合 S⋆,只发布 W′。攻击者还有假定包含这些概念的词表,但不知擦了谁、擦了几个、用了什么算法。

    Paper analysisRead analysis
  18. 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
  19. DefensesarXiv 2610.04763

    UT Austin 将 SynthID 水印嵌入基因组语言模型,凭 DNA 序列和密钥追溯 AI 生成片段

    把 SynthID 嵌入基因组语言模型采样,并用位置无关检测追溯生成序列

    Published
    Evaluated models
    Carbon-500M, GENERator-v2-eukaryote-1.2b-base
    Abstract两个六碱基基因组模型嵌入 SynthID 后,位置无关验证能找回水印,作者称这只是统计检测。

    作者在两个六碱基基因组模型的采样里嵌入 SynthID,并让验证者在不知道起点、链和 token phase 时做一次校正后的检出。

    Paper analysisRead analysis