AI RESEARCH
IndexCache: Accelerating Sparse Attention via Cross-Layer Index Reuse
arXiv CS.LG
•
ArXi:2603.12201v1 Announce Type: cross Long-context agentic workflows have emerged as a defining use case for large language models, making attention efficiency critical for both inference speed and serving cost. Sparse attention addresses this challenge effectively, and DeepSeek Sparse Attention (DSA) is a representative production-grade solution: a lightweight lightning indexer selects the top-k most relevant tokens per query, reducing core attention from $O(L^2)$ to $O(Lk.