Papers
arxiv:2507.23083

Context-aware Rotary Position Embedding

Published on Jul 30
Authors:
,
,

Abstract

CARoPE, a context-aware generalization of Rotary Positional Embeddings, enhances Transformer models by dynamically generating head-specific frequency patterns based on token embeddings, leading to improved performance and efficiency.

AI-generated summary

Positional encoding is a vital component of Transformer architectures, enabling models to incorporate sequence order into self-attention mechanisms. Rotary Positional Embeddings (RoPE) have become a widely adopted solution due to their compatibility with relative position encoding and computational efficiency. However, RoPE relies on static, input-independent sinusoidal frequency patterns, limiting its ability to model context-sensitive relationships. In this work, we propose CARoPE (Context-Aware Rotary Positional Embedding), a novel generalization of RoPE that dynamically generates head-specific frequency patterns conditioned on token embeddings. This design introduces token- and context-sensitive positional representations while preserving RoPE efficiency and architectural simplicity. CARoPE computes input-dependent phase shifts using a bounded transformation of token embeddings and integrates them into the rotary mechanism across attention heads. We evaluate CARoPE on the FineWeb-Edu-10B dataset using GPT-2 variants trained on next-token prediction tasks. Experimental results show that CARoPE consistently outperforms RoPE and other common positional encoding baselines, achieving significantly lower perplexity, even at longer context lengths. Additionally, CARoPE enables faster training throughput without sacrificing model stability. These findings demonstrate that CARoPE offers a scalable, expressive, and efficient upgrade to existing positional encoding strategies in Transformer models.

Community

Sign up or log in to comment

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2507.23083 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2507.23083 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2507.23083 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.