Skip to content

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

Oct 12 · Monday

All papers

Found 2 papers
Filter2

Model risks

Deception and scheming10Reward hacking5Misalignment and value divergence19Dangerous agent actions9Dangerous capabilities2Inappropriate refusal3Hallucination and fabrication29Bias and discrimination6

Impact

Harmful content3Output manipulation2Increased misuse capability14
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 305papers have no tags yet and are not included in filtered results.
Model risks：Dangerous capabilitiesImpact：Output manipulationClear all

Another 305papers do not have category tags yet and are not included in filtered results.

  1. Evaluation and benchmarksarXiv 2610.09964

    研究比较连续训练阶段对大语言模型说服力的影响

    比较连续后训练阶段如何改变语言模型的说服力

    arXiv
    2610.09964
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated, GPT-4
    Abstract作者称说服性 SFT 提高失调模型的态度变化，IPO 无额外收益，GPT-4 不异于中性文本。

    作者用同一人类协议，把 LLM 说服力拆成阴谋数据上的失调 SFT、说服数据上的 SFT，以及其后的 IPO。

    Paper analysisRead analysis
  2. Evaluation and benchmarksarXiv 2606.18936

    SciRisk-Bench：面向 AI4Science 安全的风险维度感知基准

    提出 SciRisk-Bench，按风险维度与学科评测 AI4Science 安全

    arXiv
    2606.18936
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    kimi-k2.6, glm-5.1, gemini-3-flash-preview and others14
    AbstractSciRisk-Bench 以风险维度和学科诊断 AI4Science 安全。

    SciRisk-Bench 是一个按风险维度和科学学科组织的 AI4Science 安全基准，用来看清失败来自哪一种风险机制、出现在哪个学科情境。作者认为 LLM 已介入科研问答、文献分析、实验规划和自主发现，而不安全输出不限于事实错误。现有科学能力基准不测安全，领域安全基准又多集中在化学、医学或生物，或只按宽泛安全类别打分。

    Paper analysisRead analysis
You have reached the end