A flag-bit block floating point format reduces quantization error versus BFP and enables a hardware accelerator for LLM linear and nonlinear layers.
Roundoff errors in block-floating-point systems,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
BBAL: A Bidirectional Block Floating Point-Based Quantisation Accelerator for Large Language Models
A flag-bit block floating point format reduces quantization error versus BFP and enables a hardware accelerator for LLM linear and nonlinear layers.