Unverified50% confidenceTradeoffExact time
Transformer的知识高度分布式编码，一个事实散布在数百万参数中，使增量更新难以精准定位存储特定旧知识的参数
1
Sources
50%
Confidence
Long-term
Relevance
7/17/2026
First Seen
Sources
Related Claims
UnverifiedTransformer由Vaswani等人于2017年提出，其知识以高度分布式方式编码，导致增量更新时无法精准定位存储特定知识的参数83% similarUnverifiedTransformer的因果注意力掩码保证每个位置只能看到历史信息，与金融预测不能使用未来数据的原则天然契合66% similarUnverified欺骗是模型在内部已表征正确信息的情况下仍输出误导性内容，是系统性偏离而非能力不足65% similarUnverifiedContinuously correcting the model stuffs more contradictory and redundant information into the context, making the model more confused, and corrections often backfire.62% similarUnverifiedFine-tuning burns knowledge into model weights, making knowledge updates costly as they require retraining; RAG stores knowledge in an external database, so updates only require modifying the database without changing the model.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/544270API
curl https://kongchang.com/api/v1/knowledge/claims/544270MCP
get_claim(id=544270)