[KongchangAI]
Unverified50% confidenceOpinionExact time

蒸馏训练范式导致开源模型学习到经过高度过滤、几乎不包含失败和纠错过程的数据分布,使模型形成不愿自我纠正的性格

1
Sources
50%
Confidence
Long-term
Relevance
7/6/2026
First Seen

Sources

Related Claims

Cite This Claim

Stable URI
https://kongchang.com/claim/114683
API
curl https://kongchang.com/api/v1/knowledge/claims/114683
MCP
get_claim(id=114683)