Unverified50% confidenceFactExact time
The typical approach in SSOD is a teacher-student framework, in which the teacher model generates pseudo boxes for unlabeled data, the student model learns on labeled data and high-confidence pseudo labels, and the teacher is updated via EMA
1
Sources
50%
Confidence
Long-term
Relevance
9/12/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedKnowledge distillation transfers the full probability distribution information output by the teacher model via soft labels, producing an implicit data augmentation effect.65% similarUnverifiedAn RSS database consisting primarily of human-original content can serve as a high-quality corpus source for AI model training, avoiding the model cannibalization problem.63% similarVerified模型蒸馏是一种将大型教师模型能力迁移到更小学生模型的技术，可能被用于扩散危险能力60% similarUnverifiedT-P-E框架暗示了机器学习的三大范式：E含输入-输出标签对应监督学习，E仅含输入无标签对应无监督学习，E以环境奖励信号出现对应强化学习60% similarUnverified模型对 schema 的遵循能力受训练数据分布影响，训练语料中高频出现的字段组合会导致模型产生幻觉字段59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/912084API
curl https://kongchang.com/api/v1/knowledge/claims/912084MCP
get_claim(id=912084)