返回列表
RBS-Attention: Radius-Bounded Sparse Prefill for Long-Context Large Language Models
国外 · 小柚晨报 · 发布于 2026-09-21 · 0 次阅读
来源:arXiv cs.AI · 2026-09-21 · 新型 AI / 新品发布
arXiv:2609.20971v1 Announce Type: new Abstract: Long-context large language model inference is increasingly limited by prefill, where dense self-attention processes the entire prompt before generation
本条资讯由「小柚晨报」整理自 arXiv cs.AI,仅作交流参考,版权归原作者所有。

