Prefill-Free Cross-Family KV Cache Transfer for Heterogeneous Multi-Agent LLMs
Results show that HeteroFold enables efficient cross-family KV reuse without receiver prefill, and achieves the best cache-transfer performance on all four long-context benchmarks and most short-context settings.
Vincent-Daniel Yun, Woo-Sang Lim, Haneul Yoo et al.
· 0 citations