Omni-Masked Gradient Descent: Memory-Efficient Optimization via Mask Traversal with Improved Convergence

Documenti analoghi

• RiskPO: Risk-based Policy Optimization via Verifiable Reward for LLM Post-Training
  di: Ren, Tao, et al.
  Pubblicazione: (2025)
• Learning Provably Improves the Convergence of Gradient Descent
  di: Song, Qingyu, et al.
  Pubblicazione: (2025)
• An Algebraically Converging Stochastic Gradient Descent Algorithm for Global Optimization
  di: Engquist, Björn, et al.
  Pubblicazione: (2022)
• FlashMask: Efficient and Rich Mask Extension of FlashAttention
  di: Wang, Guoxia, et al.
  Pubblicazione: (2024)
• Stochastic Approximation Methods for Distortion Risk Measure Optimization
  di: Jiang, Jinyang, et al.
  Pubblicazione: (2025)