RotateKV: Accurate and Robust 2-Bit KV Cache Quantization for LLMs via Outlier-Aware Adaptive Rotations
Paper
•
2501.16383
•
Published
zunhaisu
zunhai
·
AI & ML interests
None yet
Recent Activity
authored
a paper
about 1 month ago
RotateKV: Accurate and Robust 2-Bit KV Cache Quantization for LLMs via
Outlier-Aware Adaptive Rotations
authored
a paper
about 1 month ago
AKVQ-VL: Attention-Aware KV Cache Adaptive 2-Bit Quantization for
Vision-Language Models
authored
a paper
about 1 month ago
LongCat-Flash Technical Report
Organizations
None yet
authored
4
papers
about 1 month ago
authored
a
paper
5 months ago