attentive-convolution

Tag

Cards List
#attentive-convolution

RoVE: Rotary Value Embeddings Attention for Relative Position-dependent Value Pathways

arXiv cs.LG · 2026-06-11 Cached

This paper proposes RoVE, a parameter-free modification to Rotary Position Embeddings that makes value pathways position-sensitive by rotating values simultaneously with keys, transforming RoPE attention into attentive convolution. Experiments on GPT-2 models show consistent gains in few-shot in-context learning, out-of-distribution perplexity, and long-context retrieval.

0 favorites 0 likes
← Back to home

Submit Feedback