Skip to content

Reading Notes

Table of Contents

English
简体中文

Initializing search

Reading Notes

Intro
Linear Attention
Linear Attention
- 0. flash attn
  0. flash attn
  - The Hidden Recurrence in Softmax Attention: A Path to Flash Attention
- 1. retnet
  1. retnet
  - Reading Notes: RetNet
- 2. gated linear attention
  2. gated linear attention
  - Reading Notes: Implementation Notes for GLA in FLA
  - Reading Notes: Gated Linear Attention
- 3. deltanet
  3. deltanet
- 4. gated deltanet
  4. gated deltanet
- 5. kimi linear
  5. kimi linear
- 6. DPLR
  6. DPLR
  - Reading Notes: Diagonal Plus Low Rank
Multimodal Model
Multimodal Model
- Doc model
  Doc model
  - A Naive Design of Interleaved Multimodal Models for Document Understanding and Generation

Table of contents

2026.05

Table of Contents¶

2026.05¶

A Naive Design of Interleaved Multimodal Models for Document Understanding and Generation

Comments

Made with Material for MkDocs