cd/entity/Multi-Head Latent Attention· home› entities› Multi-Head Latent Attention

grep -l @multi-head latent attention /news/*.json | wc -l → 3

Multi-Head Latent Attention

mentions 3 type Person feed RSS

// recent coverage 3 mentions

02:49

2026-07-21

arxiv.org

artificial-intelligence

Kimi Linear: An Expressive, Efficient Attention Architecture

Researchers at Moonshot AI introduced Kimi Linear, a hybrid linear attention architecture that outperforms full attention across short-context, long-context, and reinforcement learning scaling regimes…

14:15

2026-06-25

dev.to

large-language-models

Why KV Cache Matters — How MQA, GQA, and MLA Make LLM Inference Faster

KV Cache reduces duplicated computation in autoregressive LLM inference by storing previously computed Key and Value tensors, but creates a memory bottleneck as context length grows. To address this, …

13:14

2026-05-23

dev.to

large-language-models

Multi-Head Latent Attention (MLA)

**Summary:** Multi-Head Latent Attention (MLA) is an attention mechanism used in DeepSeek-V2/V3 and Kimi K2.x models that compresses the Key-Value (KV) cache by projecting full KV pairs into a shared,…

// co-occurs with top 8 entities

Multi-Head Attention 2 DeepSeek-V2 1 DeepSeek-V3 1 Kimi K2.x 1 KV Cache 1 Multi-Query Attention 1 Grouped-Query Attention 1 Moonshot AI 1

// topics top 6 topics

large language models 3 artificial intelligence 2 machine learning 2 ai infrastructure 2 research 1 ai research 1