Skip to content

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

Oct 12 · Monday

All papers

Found 1 paper
Filter3

Defense type

Model hardening1Inference-time intervention6Detection and filtering9Instruction–data separation11Permissions and approval14Monitoring and auditing58

System component

Memory0Guardrails and judges11Monitors1Reasoning chains11Fine-tuning and open weights18Vision12Audio5Image generation5Multimodal22
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.
Defense type：Model hardeningSystem component：MemorySystem component：MonitorsClear all

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

  1. DefensesarXiv 2609.38645

    用探针引导微调对齐模型且不损失可监控性

    用持续更新的探针做微调，降低有害性并保持线性可监测性

    arXiv
    2609.38645
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    Mistral 7B Instruct v0.1, Llama 3 8B Instruct Abliterated, Qwen3-14B
    Abstract持续重拟合的探针可降低有害性、提高诚实性，训练后概念仍能被线性探针读出。

    Probe-guided fine-tuning 把探针分数当作唯一对齐信号，用来推动 harmlessness 与 honesty，并检查训练后这些概念是否仍线性可读。

    Paper analysisRead analysis
You have reached the end