HASTE: Hardware-Aware Dynamic Sparse Training for Large Output Spaces

Documenti analoghi

• DynaSpec: Context-aware Dynamic Speculative Sampling for Large-Vocabulary Language Models
  di: Zhang, Jinbin, et al.
  Pubblicazione: (2025)
• Navigating Extremes: Dynamic Sparsity in Large Output Spaces
  di: Ullah, Nasib, et al.
  Pubblicazione: (2024)
• ELMO: Efficiency via Low-precision and Peak Memory Optimization in Large Output Spaces
  di: Zhang, Jinbin, et al.
  Pubblicazione: (2025)
• Labels in Extremes: How Well Calibrated are Extreme Multi-label Classifiers?
  di: Ullah, Nasib, et al.
  Pubblicazione: (2024)
• Large Language Model as a Teacher for Zero-shot Tagging at Extreme Scales
  di: Zhang, Jinbin, et al.
  Pubblicazione: (2024)