---
id: 20260716-T0-08
title: "SFDA：一种新型线性注意力机制，用傅里叶相位控制记忆并提升长上下文"
title_en: "SFDA: Novel Linear Attention Using Fourier Phase Control for Better Long-Context Memory"
url: https://ai.daily.yangsir.net/daily/20260716-T0-08
issue_date: 2026-07-16
publish_date: 2026-07-15T04:00:00.000Z
category: research
source_name: "arXiv cs.LG (ML)"
source_url: https://arxiv.org/abs/2607.11897
---

# SFDA：一种新型线性注意力机制，用傅里叶相位控制记忆并提升长上下文

arXiv 新论文提出半直接傅里叶Delta注意力（SFDA），通过相位控制的Delta记忆和构造性Chunk-WY内核，解决了线性注意力中固定循环状态导致的长上下文记忆丢失问题。该方法在不增加KV缓存的前提下，提升了模型对精确状态追踪和长序列处理的能力。

## English Version

**SFDA: Novel Linear Attention Using Fourier Phase Control for Better Long-Context Memory**

A new paper introduces Semidirect Fourier Delta Attention (SFDA), which uses phase-controlled delta memory and constructive Chunk-WY kernels to overcome long-context memory loss in linear attention. It improves state tracking and long-sequence processing without increasing KV cache size.

---

**来源**：[arXiv cs.LG (ML)](https://arxiv.org/abs/2607.11897)

**详情页**：https://ai.daily.yangsir.net/daily/20260716-T0-08

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*