arXiv Machine Learning By Shun-ichiro Hayashi, Daichi Mukunoki, Tetsuya Hoshino, Takahiro Katagiri

ATFlash: Per-RoPE-Wavelength Attention Windows for Compute/Memory-Efficient LLM Inference

Read the original on arXiv Machine Learning →

arXiv:2608. 02947v1 Announce Type: new Abstract: The attention score with rotary position embeddings (RoPE) decomposes exactly into a sum over its 2D-rotation frequency pairs, and each pair's wavelength limits how far it can discriminate position.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.

Hugging Face Trending Papers
Aug 3

ATFlash: Per-RoPE-Wavelength Attention Windows for Compute/Memory-Efficient LLM Inference

The attention score with rotary position embeddings (RoPE) decomposes exactly into a sum over its 2D-rotation frequency pairs, and each pair's wavelength limits how far it can discriminate position. Aligned with this structure, we propose the per-RoPE-wavelength distance window: it prunes the query--key inner-product terms beyond a wavelength-proportional distance.

arXiv AI
Jul 10

Jet-Long: Efficient Long-Context Extension with Dynamic Bifocal RoPE

arXiv:2607. 07740v1 Announce Type: cross Abstract: Modern LLMs are increasingly deployed in long-context applications such as retrieval-augmented generation, repository-level coding, and agentic workflows whose accumulated reasoning and tool traces routinely push the input an order of magnitude past the pretraining window, making zero-shot context extension the dominant deployment path for open-weight checkpoints.

By Haozhan Tang, Zerui Wang, Yuxian Gu, Song Han, Han Cai