Unverified50% confidenceOpinionExact time
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
1
Sources
50%
Confidence
Long-term
Relevance
7/6/2026
First Seen
Sources
Adding a 'Repair Layer' to Open-Source Models: How DeepSeek Can Outperform Claude Opus
bilibiliGelai_AI7/5/2026
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.82% similarUnverified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69% similarUnverifiedWhen a model lacks accurate knowledge of a particular library, it tends to generate syntactically correct but semantically incorrect code—so-called 'confident errors.'69% similarUnverifiedThe author believes that the poor tool-calling performance of open-source models is in the vast majority of cases a problem with the framework, not a deficiency in the model's own capabilities.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.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/114162API
curl https://kongchang.com/api/v1/knowledge/claims/114162MCP
get_claim(id=114162)