85%Must read
?Must readVote to see the score
GradientStabilizer:Fix the Norm, Not the Gradient
GradientStabilizer replaces gradient magnitudes with running-statistic estimates to eliminate spike-induced divergence across LLM pretraining, quantization-aware training, and reinforcement learning.
Published Feb 24, 2025 · 0 citations · ▲ 20 on Hugging Face · Code ★ 28
– Readers
14/20 AI panelreviewers recommend it
Readers and the AI panel: vote on this paper to see what they said.
Only vote on papers you've read. Sign in to vote.
AI panel: 14 of 20 reviewers recommend it
lenient 5/5
medium 8/10
strict 1/5