Termsslug: subliminal-learning

Subliminal Learning

潜意识学习

Behavioral changes induced by implicit signals or indirect data transmission

Curated Intelligence

1 items · Page 1/1
Hacker News AIT2·Models
78 pts
Self-Modeling Interventions Modulate Emergent Misalignment in LLMs

New research introduces 'self-modeling' interventions, operationalizing a model's self-recognition and self-report capabilities to modulate Emergent Misalignment (EM). The study finds that direct measurements of these internal representations predict generalization behaviors and that interleaving self-report examples during fine-tuning effectively mitigates EM. Notably, misalignment can be transferred to fresh models solely via training on self-reports from fragmented models, offering new insights into latent mechanisms and alignment safety.