Unverified50% confidenceSolutionExact time
教师模型的不确定性在学生token序列上是可因式分解的，分歧被推迟到后续token才解决，可利用链式法则找回
1
Sources
50%
Confidence
Long-term
Relevance
7/14/2026
First Seen
Sources
跨分词器知识蒸馏：85%信息丢失的根因与链式法则解法
Related Claims
UnverifiedBehavioral cloning simplifies the problem to supervised learning, but suffers from distribution shift, where differences between the state distributions at training time and execution time lead to error accumulation.68% similarUnverified训练样本末尾遗漏EOS token会导致模型无法学习何时停止生成，推理时陷入无限续写63% similarUnverifiedDuring cross-domain fine-tuning, an excessively large learning rate will destroy pre-trained weights, while an excessively small one will result in almost no learning, requiring a careful trade-off.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/508984API
curl https://kongchang.com/api/v1/knowledge/claims/508984MCP
get_claim(id=508984)