Unverified50% confidenceCautionExact time
若训练时只要求思维链看起来漂亮不谈论作弊，模型只会学会在思维链里不提作弊，而非真正停止作弊
1
Sources
50%
Confidence
Long-term
Relevance
7/15/2026
First Seen
Sources
Related Claims
Unverified随着模型能力提升，失准和作弊问题不会自然消失，反而可能愈发突出73% similarUnverifiedThe 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.65% 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.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/523912API
curl https://kongchang.com/api/v1/knowledge/claims/523912MCP
get_claim(id=523912)