Unverified50% confidenceFactExact time
Decoupled exponential moving average (D-EMA) adopts a configurable grouped averaging strategy that constrains interference-sensitive pretrained parameters while preserving flexibility for other parameters.
1
Sources
50%
Confidence
Long-term
Relevance
10/15/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedExponential moving average (EMA) is a weight smoothing technique used in training that maintains a sliding-average copy of the parameters and is insensitive to the noise and abrupt changes of single-step gradients.79% similarVerifiedAlignment algorithms such as DPO (Direct Preference Optimization) reduce training costs while improving the consistency of model performance on open-ended instructions.75% similarUnverified后训练优化包括指令微调、RLHF或DPO策略调整、系统提示词工程改进及工具调用和多模态融合能力强化69% similarUnverifiedOn infilling tasks, Diffusion-LM outperforms continuous-relaxation-based methods such as COLD and DELOREAN on automatic evaluation metrics, and reaches the level of autoregressive models trained from scratch specifically for this purpose without requiring targeted training.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1037827API
curl https://kongchang.com/api/v1/knowledge/claims/1037827MCP
get_claim(id=1037827)