Unverified50% confidenceBenchmarkExact time
Under long training cycles, MuonGS outperforms MuonH on validation perplexity.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/28/2026
First Seen
Valid until: 1/26/2027
Sources
Related Entities
Related Claims
UnverifiedThe paper proposes an optimizer method called MuonGS that accelerates LLM pretraining by relaxing weight norm control.74% similarUnverifiedZGCM-1 uses a stable FP8 Muon optimizer, combining the Muon optimizer with FP8 low-precision training while maintaining training stability.72% similarUnverifiedThe original Muon optimizer can only achieve logarithmic growth in the cumulative effective learning rate during training, which may limit the model's exploration capability.70% similarUnverifiedMUON优化器替代了沿用多年的AdamW优化器，通过引入矩阵正交化步骤使训练曲线更加平滑68% similarUnverified强化学习方案通过在MuJoCo、Isaac Gym等物理仿真引擎中进行数亿次试错交互训练控制器，仿真训练可将时间压缩数百倍66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1106002API
curl https://kongchang.com/api/v1/knowledge/claims/1106002MCP
get_claim(id=1106002)