Showing cs.CLShow all
2 papers · 1 filter
cs.CL2026
Positions Are Not Facts: The Mismatch Between KV Caches and Memory
Changhai Zhou, Yuhua Zhou, Shiyang Zhang +5
When a fact changes, how should a language model update the history stored in its key-value (KV) cache? Hiding the old record is cheap, but it may still contain needed details or a…
cs.CL2024
RankAdaptor: Hierarchical Rank Allocation for Efficient Fine-Tuning Pruned LLMs via Performance Model
Changhai Zhou, Shijie Han, Lining Yang +4
The efficient compression of large language models (LLMs) has become increasingly popular. However, recovering the performance of compressed LLMs remains a major challenge. The cur…