Skip to content

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

Oct 12 · Monday

All papers

Found 2 papers
Filter2

Impact

Output manipulation2

System type

Models and APIs4LLM applications2
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 265papers have no tags yet and are not included in filtered results.
Impact：Output manipulationSystem type：LLM applicationsClear all

Another 265papers do not have category tags yet and are not included in filtered results.

  1. OtherarXiv 2607.21317

    研究称全球南方 AI 审计远落后于部署，十年仅不足二十例已部署系统审计

    论证全球南方已部署 AI 系统的独立审计远落后于部署

    arXiv
    2607.21317
    Published
    Layer
    Application layer
    Setting
    LLM applications
    Evaluated models
    Robot Laura, national child-welfare risk model (Southern Cone, anonymized), national public employment matching algorithm (Latin America, anonymized) and others4
    Abstract作者用自身委托记录评估缺口和四类模式，并主张资助方把独立评估写成用款条件。

    Eticas 的立场是：全球南方的 AI 已经进入公共系统，独立评估还没有成为义务，先能动的是出资方。

    Paper analysisRead analysis
  2. OtherarXiv 2605.22949

    MARGIN：面向多智能体基础模型协调的运行时置信度校准

    提出 MARGIN 在线校正异构模型自报置信并用于答案投票

    arXiv
    2605.22949
    Published
    Layer
    Application layer
    Setting
    LLM applications
    Evaluated models
    MiniMaxAI/MiniMax-M2.5, Qwen/Qwen2.5-14B-Instruct-1M, Qwen/Qwen2.5-32B-Instruct and others18
    Abstract有反馈时，MARGIN 按模型校正自报置信，并在两次代码迁移上得到更低 ECE。

    MARGIN 是放在协调器侧的运行时置信校正：异构基础模型各自报告置信，协调器不改模型，只把这些分数变成答案级投票的权重。要处理的是两件事。同一自报数值在不同模型上可靠性不同。

    Paper analysisRead analysis
You have reached the end