Unverified50% confidenceFactExact time
LoRA decomposes the low-rank adaptation matrix into A and B, where B is initialized to zero to ensure that the adaptation contribution is zero at initialization.
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
Unverified从零实现 Adam 需要处理偏差校正（bias correction），因为初始阶段 m 和 v 被初始化为零导致估计偏低66% similarUnverifiedRidge regression shrinks regression coefficients toward zero but does not set them to zero by adding an L2 regularization term to the loss function, whereas LASSO (L1 regularization) produces sparse solutions.66% similarUnverifiedLoRA decomposes the fine-tuning weight change ΔW into the product of two low-rank matrices, with a parameter count of only 0.1%–1% of the original model, achieving a parameter compression ratio as high as 256× when r=8 and d=4096.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/893770API
curl https://kongchang.com/api/v1/knowledge/claims/893770MCP
get_claim(id=893770)