Hybrid linear attention plus ternary projections, co-designed with a multiplier-free PE, deliver 10× weight and 12.8× KV-cache compression with competitive MMLU and FPGA speedups over LLaMA 3 on A100.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AR 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
ELiTeFormer: An Efficient Transformer for FPGAs
Hybrid linear attention plus ternary projections, co-designed with a multiplier-free PE, deliver 10× weight and 12.8× KV-cache compression with competitive MMLU and FPGA speedups over LLaMA 3 on A100.