推荐阅读，@OwainEvans_UK 团队的所有文章篇篇精彩！
Beyond Owls:
Subliminal Learning Can Transfer Learned Capabilities and Backdoors
Subliminal learning transfers capabilities and backdoors between models
AISummary
Owain Evans's team reports that models can transfer more complex traits than owl preferences through training data, including novel skills, agentic hacking, and backdoors. In the new paper, a backdoor transfers even though neither the trigger nor the backdoor behavior appears in the training data.
Post on XView on X
Dongxi NLPVerified on X
@dongxi_nlp
New paper. Previous: Models transfer a love of owls through sequences of numbers (Subliminal Learning). New: We show models can transfer more complex traits: novel skills, agentic hacking, backdoors. Here, a backdoor transfers with neither trigger nor behavior in the data!
Source: Dongxi NLP · x.comPublished · added here