Skip to content

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

Oct 13 · Tuesday

All papers

Found 1 paper
Filter5

Attack type

Prompt injection14Poisoning and backdoors8Detection evasion1

System type

LLM applications1coding agent0

Attack surface / entry point

User input13Documents and messages3Tool and skill descriptions3Code repositories and dependencies1Memory writes0
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：Detection evasionSystem type：LLM applicationsSystem type：coding agentAttack surface / entry point：Code repositories and dependenciesAttack surface / entry point：Memory writesClear all

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

  1. DefensesarXiv 2608.06422

    CMU 研究：分片判断可避免 LLM 监督失效并抵御对抗利用

    提出分片监督，将评判标准分配给独立调用以抵御展示攻击

    arXiv
    2608.06422
    Published
    Layer
    Application layer
    Setting
    LLM applications
    Attacker
    Black-box
    Evaluated models
    Claude Opus 4.8, Claude Sonnet 4.6, Claude Opus 4.6 and others4
    Abstract论文揭示决策负载是大模型监督的独立瓶颈，提出分片与辩论机制有效分离注意力与平衡证据，在多项高难度基准上实现稳健评判。

    本文系统研究了大语言模型在执行多标准监督与评判时的决策负载瓶颈及其安全风险。

    Paper analysisRead analysis
You have reached the end