HySparse2: Hybrid Sparse Attention with Two-Level KV Sharing
Long-horizon and multi-turn agents typically generate short actions and process long observations from tools and environments. This growing context demands efficient prefill, compact KV-cache storage, and accurate long-context retrieval. To meet these demands, we introduce HySparse2, a hybrid sparse attention architect...
Jian-Yu Wei, Yi-Zhao Gao, Qi-Hao Zhang et al.
· 0 citations