Skip to content

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

Oct 12 · Monday

Reward hacking · Papers

Found 4 papers
Filter1

Model risks

Reward hacking4

Impact

Oversight failure3

Layer

Application layer3

System component

Guardrails and judges4Monitors5Reasoning chains3
Select any value within a category; different categories must all match. Counts include the other filters you have selected.
System component：Guardrails and judgesTopic: Reward hackingClear all
  2. Evaluation and benchmarksarXiv 2609.34262

    论文提出基准完整性维护框架，审计 Agent 通过轨迹中的非应得通过

    审计 Agent 基准通过轨迹，区分无根据通过与奖励黑客并封堵复测

    arXiv
    2609.34262
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Opus 4.7, Opus 4.8, Opus 5 and others11
    Abstract审计通过轨迹、分开奖励黑客与验证器弱点，并用重放加现场重测验收修复。

    Scale AI 的这篇工作把 agentic benchmark 的有效性写成持续维护问题：验证器给满分，并不等于智能体展示了任务所要的能力。

    Paper analysisRead analysis
  3. Risky behaviorarXiv 2606.04923

    CHERRL：可复现、分析与检测评分标准强化学习中的奖励作弊

    提出 CHERRL 注入裁判偏差以复现奖励作弊，并用 RHDA 检测训练起点

    arXiv
    2606.04923
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    Qwen3-4B, Qwen3.5-plus, Qwen3.5-397B-A17B and others4
    AbstractCHERRL 注入已知偏差以观察奖励黑客，并用于分析偏差和评测起点检测器 RHDA。

    CHERRL 是 rubric-based RL 的可控环境，用来复现奖励黑客、观察奖励分歧，并给检测器提供起点参照。

    Paper analysisRead analysis
You have reached the end