Waterfilling rate allocation makes quantized matrix multiplication for LLMs near information-theoretically optimal, with WaterSIC being basis-free and within 0.25 bits per entry of the limit.
On lovász’lattice reduction and the nearest lattice point problem,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
High-Rate Quantized Matrix Multiplication II
Waterfilling rate allocation makes quantized matrix multiplication for LLMs near information-theoretically optimal, with WaterSIC being basis-free and within 0.25 bits per entry of the limit.