# HyScale: Scaling Long-Context Training for Learned Sparse Attention

> Толық мәтін ағылшын тілінде

**Published:** 2026 ж. 26 қазан
**Author:** Yangsheng Deng, Xinjing Huang, Yulong Wang, Yunzhuo Liu, Jun Hu, Shuyong Tang, Junchi Zeng, Mingji Han, Ray Ying, Key Zhang, Luo Mai
**Source:** arXiv — cs.AI
**Language:** en-US
**HTML:** https://aiqazaqstan.com/kk/articles/hyscale-scaling-long-context-training-for-learned-sparse-attention

arXiv:2610.32057v1 Announce Type: new Abstract: We present HyScale, a long-context training system for learned sparse attention. Learned sparse attention is a promising architecture for million-token contexts, widely adopted in production since DeepSeek-V3.2. Selection, however,…

> Материал бастапқы дереккөзден алынған: **arXiv — cs.AI**
> [Толық мәтінді arXiv — cs.AI сайтында оқу](https://arxiv.org/abs/2610.32057)

**Тегтер:** #arXiv #DeepSeek

---

*AI Qazaqstan — non-commercial informational portal about AI in Kazakhstan. Machine translations and summaries are AI-generated.*