FastGen adaptively compresses LLM KV caches via lightweight attention profiling: evicting long-range contexts on local heads, non-special tokens on special-token heads, and retaining full caches on broad-attention heads, yielding substantial memory savings with negligible quality loss.
Structured Pruning Learns Compact and Accurate Models , booktitle =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2023 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs
FastGen adaptively compresses LLM KV caches via lightweight attention profiling: evicting long-range contexts on local heads, non-special tokens on special-token heads, and retaining full caches on broad-attention heads, yielding substantial memory savings with negligible quality loss.