FineQ quantizes LLM weights to 2.33 bits by protecting outliers inside 3-weight clusters and adds a temporal-coding accelerator that replaces multipliers with adders.
ubrain: A unary brain computer interface,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
FineQ: Software-Hardware Co-Design for Low-Bit Fine-Grained Mixed-Precision Quantization of LLMs
FineQ quantizes LLM weights to 2.33 bits by protecting outliers inside 3-weight clusters and adds a temporal-coding accelerator that replaces multipliers with adders.