CyclicFlowAttention - a cyxxxxxxxxxx Collection

cyxxxxxxxxxx 's Collections

CyclicFlowAttention

CyclicFlowAttention

updated about 5 hours ago

Pretrained Cyclic Flow Attention (CyFA) language models.

cyxxxxxxxxxx/cyfa-400M-15B

Text Generation • 0.4B • Updated about 5 hours ago • 340
cyxxxxxxxxxx/cyfa-800M-30B

Text Generation • 0.8B • Updated about 5 hours ago • 324
cyxxxxxxxxxx/cyfa-1.4B-100B

Text Generation • 1B • Updated about 5 hours ago • 303
CyFA: Linear Sequence Modeling with Relative-Time-Partitioned Memory

Paper • 2609.36259 • Published 3 days ago

Note Cyclic Flow Attention (CyFA)