Gated-delta network attention-sink free architecture
Apr 06 , 2026
Parallelizing recurrence with DeltaNet
Apr 06 , 2026
The Transformer and the Cortex: A Study in Parallel Design
Apr 05 , 2026
Attention Sink Problem in Transformer Architecture
Apr 05 , 2026
Vectors- a mathematical concept that rules modern AI
Apr 05 , 2026