Alibaba Cloud Big Data AI Platform
Dec 23, 2025 · Artificial Intelligence
How Skrull Boosts Long-Context Fine‑Tuning Speed Up to 7.5×
The Skrull system, accepted at NeurIPS 2025, dynamically schedules long and short sequences during each training iteration, overlapping communication and computation to achieve up to 7.54× speedup for long‑context fine‑tuning of large language models while maintaining stability through load‑balancing and rollback mechanisms.
Dynamic Data SchedulingLarge Language ModelsLong Context Fine-Tuning
0 likes · 8 min read