Skip to content

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

Oct 12 · Monday

All papers

Found 1 paper
Filter4

Attack type

Extraction and theft1

Layer

Model layer3Training and data layer1

System component

Memory0Reasoning chains1Fine-tuning and open weights3Multimodal3
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 254papers have no tags yet and are not included in filtered results.
Attack type：Extraction and theftLayer：Training and data layerSystem component：MemorySystem component：Reasoning chainsClear all

Another 254papers do not have category tags yet and are not included in filtered results.

  1. AttacksarXiv 2609.35699

    论文：蒸馏防御在蒸馏后强化学习下易被攻破

    提出用 API 推理摘要重建轨迹再蒸馏加强化学习，攻破蒸馏防御

    arXiv
    2609.35699
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    Qwen2.5-0.5B, Qwen2.5-1.5B, Qwen2.5-3B and others12
    Abstract更现实的蒸馏攻击是蒸馏后再 RL。

    这篇工作修订蒸馏攻击的威胁模型：攻击者经合法 API 收集闭源推理数据，先蒸馏自有模型，再做强化学习。作者认为只在蒸馏后评估防御，会漏掉后续 RL，从而高估防御。

    Paper analysisRead analysis
You have reached the end