cd/entity/Rotary Position Embedding· home› entities› Rotary Position Embedding
grep -l @rotary position embedding /news/*.json | wc -l → 1

Rotary Position Embedding

mentions 1 type Person feed RSS

// recent coverage 1 mentions

19:13
2026-07-22
dev.to
large-language-models

RoPE: How 2D Rotations Solved Transformer Long-Context

Rotary Position Embedding (RoPE), introduced by Su et al. in 2021, solves the Transformer long-context problem by rotating Query and Key vectors in 2D sub-planes, making attention depend only on relat…

// co-occurs with top 3 entities
Su et al. 1 RoPE 1 FlashAttention 1
// topics top 5 topics
large language models 1 machine learning 1 artificial intelligence 1 neural networks 1 ai research 1