You Only Index Once: Cross-Layer Sparse Attention with Shared Routing
Why This Matters
This paper is highly relevant to LLM integration, fine-tuning, and production AI. It proposes a novel cross-layer sparse attention architecture for long-context inference in large language models, demonstrating its efficiency and accuracy in certain scenarios. The paper provides specific technical content, including methods, benchmarks, and empirical results.
Abstract
Long-context inference in modern LLMs is increasingly constrained by decoding efficiency, especially in reasoning-heavy settings where models generate long intermediate chains of thought. Existing sparse attention methods often face a practical efficiency-quality trade-off. Structured block sparse methods typically provide stronger acceleration but incur noticeable quality loss, while token sparse methods are usually more accurate yet deliver limited end-to-end speedup because top-k routing over the full cache remains expensive. In this work, we propose cross-layer sparse attention (CLSA), which is built on top of KV-sharing architectures such as YOCO. The core idea is to share not only the KV cache across cross-decoder layers, but also the routing index. A single indexer computes token-level top-k selection once and reuses the resulting index across layers, thereby preserving the fine-grained selectivity of token sparse attention while amortizing the routing overhead. The resulting architecture improves all major inference bottlenecks jointly, including pre-filling, KV-cache storage, and long-context decoding. Experiments across short-context and long-context benchmarks show that CLSA is both accurate and efficient, achieving up to 7.6x decoding speedup and 17.1x overall throughput improvement at 128K context. These results suggest a more complete architectural solution for long-context LLMs that jointly advances model quality and inference efficiency.
Links
Metadata
Save to Vault
Save this article directly to your Obsidian vault. Opens Obsidian with the note pre-filled.
Will save to: vault/inbox/signals/2026-06-07-you-only-index-once-cross-layer-sparse-attention-with-shared-routing.md