A hybrid SLC/MLC RRAM accelerator with gradient-redistribution fine-tuning keeps transformer accuracy while moving most weights into denser, more efficient MLC storage.
Title resolution pending
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
-
Hybrid SLC-MLC RRAM Mixed-Signal Processing-in-Memory Architecture for Transformer Acceleration via Gradient Redistribution
A hybrid SLC/MLC RRAM accelerator with gradient-redistribution fine-tuning keeps transformer accuracy while moving most weights into denser, more efficient MLC storage.