Статьи
ИИКлючевая идея
A hidden AI bias you can't see in its text? Force it to leak.
ИИКлючевая идея
Модель, обучаясь у самой себя, теряет разнообразие решений
ИИКлючевая идея
Teaching an AI to grade its own reasoning with a checklist beats giving it one right answer
ИИУпоминается
Teaching AI to solve math by finding problems it can copy from
Смежные понятия
Идеи, которые идут в паре с понятием «Knowledge Distillation».
AI SafetyBias Detection In Language ModelsChain-Of-ThoughtCredit AssignmentFisher InformationGraph Path-FindingInformation RetrievalLanguage Model ReasoningLarge Language ModelsMathematical ReasoningModel AuditingModel Supply Chain SecurityMutual InformationOn-Policy Self-DistillationOut-Of-Distribution GeneralizationOutput DiversityPrefix TuningReasoning By AnalogyReasoning Language ModelsReinforcement Fine-TuningReinforcement LearningReinforcement Learning With Verifiable RewardsRetrieval-Augmented GenerationReward DesignRubric-Conditioned Self-DistillationScience Reasoning BenchmarksSelf-Distillation