Unverified50% confidenceFactExact time
多对一映射导致教师token间的不确定性在投影到学生第一个token时被直接加和抹平，这是投影方式的固有属性而非实现错误
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
UnverifiedThe teacher model outputs in distillation training rarely include self-correction examples, causing student models to form an implicit belief that "strong models don't make mistakes," leading them to repeat rather than correct when verification fails.72% similarUnverified许多开源模型继承教师模型的自信偏差，导致工具调用错误时不愿自我纠正69% similarUnverifiedThe root cause of rising hallucination rates in reasoning models is misaligned incentives: reinforcement learning training causes models to tend to generate detailed and persuasive chains of thought, and human evaluators give higher scores to more elaborately argued answers, even when they contain factual errors.69% similarUnverified候选人在代码第一行就写出WeakSet核心解法却对细节无法解释，这种由果推因的线性确定性不符合人类探索性认知特征，暴露了照抄AI输出的状态66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/508979API
curl https://kongchang.com/api/v1/knowledge/claims/508979MCP
get_claim(id=508979)