A data-driven adaptive policy for KV-cache bit-width selection based on token importance features reduces decoding latency by ~18% and improves accuracy over static quantization while staying near FP16 levels on SmolLM models.
H2o: Heavy-hitter ora- cle for efficient generative inference of large language mod- els.Advances in Neural Information Processing Systems, 36: 34661–34710, 2023
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Don't Waste Bits! Adaptive KV-Cache Quantization for Lightweight On-Device LLMs
A data-driven adaptive policy for KV-cache bit-width selection based on token importance features reduces decoding latency by ~18% and improves accuracy over static quantization while staying near FP16 levels on SmolLM models.