Unverified50% confidenceFactExact time
Transformer 架构的模型参数在训练完成后即被固定，除非进行微调或持续预训练否则无法获取新知识
1
Sources
50%
Confidence
Long-term
Relevance
8/23/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedCompared to RNN/LSTM architectures, Transformers can be highly parallelized during training, making it possible to train models with hundreds of billions of parameters73% similarUnverified一项研究提出仅训练单个Transformer层，其效果就能与全参数RL训练相媲美72% similarUnverifiedParametric knowledge is fixed in model weights and cannot be updated without retraining once training is complete; non-parametric knowledge is stored in external databases and can be added, deleted, or modified at any time.71% similarUnverifiedTransformer、残差连接、批归一化等突破性架构与训练技巧均先在实践中被证明有效，理论解释往往滞后数年71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/789719API
curl https://kongchang.com/api/v1/knowledge/claims/789719MCP
get_claim(id=789719)