一作ACL 2026Poster· Natural Language Processing
REAL : RE trieval-re A soning and Logic-constructed Attention Behaviors for Long-Context KV Cache Compression
Mengjie Li · yuan feng · Xike Xie · William J. Song
摘要
The growing sequence length of large language models poses significant challenges for key-value (KV) caches. Existing state-of-the-art cache eviction methods primarily analyze the inference behavior of attention heads in successful retrieval-reasoning cases, often overlooking diverse behaviors in...