Skip to content

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

Oct 13 · Tuesday

All papers

Found 6 papers
Filter4

Model risks

Deception and scheming3

Defense type

Monitoring and auditing3

Impact

Oversight failure5

System type

Models and APIs4

Layer

Model layer6Application layer34

System component

RAG0Guardrails and judges5Monitors6Reasoning chains6Vision8Audio4Video5Image generation3Embeddings0Multimodal16
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.
Layer：Model layerSystem component：RAGSystem component：MonitorsSystem component：EmbeddingsClear all

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

  2. Evaluation and benchmarksarXiv 2610.04737

    PyINE：用可验证 Python 执行轨迹研究推理模型的捷径与监督

    提出 PyINE 基准，训练捷径跟随模型并比较监督者

    arXiv
    2610.04737
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    shortcut-following model organism, gpt-4o, gpt-5 and others12
    AbstractPyINE 用可验证代码执行研究捷径失败的监督覆盖与成本权衡。

    PyINE 是用可执行程序研究推理模型引出与监督的框架，PyINE-v1 是其第一次发布。

    Paper analysisRead analysis
  3. Evaluation and benchmarksarXiv 2610.02741

    研究系统评估循环语言模型的思维链可监控性

    系统评估循环语言模型加深 loop 后的 CoT 可监控性

    arXiv
    2610.02741
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Ouro-1.4B-Thinking, Ouro-2.6B-Thinking, Nanbeige4.2-3B and others4
    Abstract加深 loop 可在部分压力测试任务降低可监测性，循环架构本身并不必然更难监测。

    这项工作评估 LoopLM 的 CoT 可监测性：加深共享层的重复次数，以及“有循环结构”本身，会不会让决策关键因素更少出现在可监测文本里。

    Paper analysisRead analysis
  4. Evaluation and benchmarksarXiv 2609.39533

    CATCH: A Controllable Analysis Testbed for Coding RL Reward Hacking

    构建编码 RL 奖励作弊测试台 CATCH 并检验 CoT 监控失效

    arXiv
    2609.39533
    Published
    Layer
    Model layer
    Setting
    coding agent
    Evaluated models
    Qwen3-4B, Qwen3.5-27B
    AbstractCATCH 用双运行金标签研究编码 RL 的奖励黑客，并观察到监控会被注释适应削弱。

    CATCH 是面向编码 RL 奖励黑客的可控分析测试床，用可利用环境、可调初始倾向和基于执行的金标签，在训练全程比较黑客动态与干预。

    Paper analysisRead analysis
  5. Evaluation and benchmarksarXiv 2602.14161

    When benchmarks lie: evaluating malicious prompt classifiers under real-world distribution shift

    提出 LODO 评测恶意提示探针的跨数据集泛化

    arXiv
    2602.14161
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Llama-3.1-8B-Instruct, Gemma-3-27B, Qwen-3.5-4B and others5
    Abstract作者称 LODO 露出 CV 高估，数据集身份与安全方向部分重叠，标准域泛化干预关不上。

    LODO 用来检查恶意提示词分类器换成没见过的数据集后，还剩多少检测能力。。

    Paper analysisRead analysis
  6. Evaluation and benchmarksarXiv 2602.08877

    Automated red-teaming pipeline stress-tests alignment audits, finds prompt-level strategic deception can render audits ineffective

    提出自动化红队流水线，用系统提示词压测对齐审计能否抵御策略性欺骗

    arXiv
    2602.08877
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Gemma 2 9B Instruct, Llama 3.3 70B Instruct

    Abstract通过红队提示词揭示现有审计方法易受策略欺骗，并在激活层面诱导高置信度错误推断。

    Paper analysisRead analysis
You have reached the end