PM-KVQ introduces progressive bit-width shrinking, per-block memory allocation, and positional-interpolation calibration to make 2-bit KV cache quantization nearly lossless on long-CoT LLMs.
Carnegie mellon informatics and mathematics competition, 2025
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
dataset 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
dataset 1polarities
use dataset 1representative citing papers
citing papers explorer
-
PM-KVQ: Progressive Mixed-precision KV Cache Quantization for Long-CoT LLMs
PM-KVQ introduces progressive bit-width shrinking, per-block memory allocation, and positional-interpolation calibration to make 2-bit KV cache quantization nearly lossless on long-CoT LLMs.