Skip to content

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

Oct 12 · Monday

Reward hacking · Papers

Found 1 paper
Filter1

Defense type

Model hardening1
Select any value within a category; different categories must all match. Counts include the other filters you have selected.
Defense type：Model hardeningTopic: Reward hackingClear all
  1. Risky behaviorarXiv 2609.39102

    论文提出 CrossFit 缓解自演化搜索智能体的共谋作弊

    提出 CrossFit，用折外求解器抑制自演化搜索智能体共作弊

    arXiv
    2609.39102
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Qwen3.5-4B, Qwen3.5-9B
    AbstractCrossFit 切断同源伪标签回流到提议者奖励的路径，并提高七个搜索基准上的 Cover-EM。

    False Frontiers 诊断自进化搜索智能体中的 co-cheating，并用按来源交叉拟合的反馈来缓解它。

    Paper analysisRead analysis
You have reached the end