AI RESEARCH
Nautile-370M: Spectral Memory Meets Attention in a Small Reasoning Model
arXiv CS.LG
•
ArXi:2604.24809v1 Announce Type: new We present Nautile-370M, a 371-million-parameter small language model designed for efficient reasoning under strict parameter and inference budgets. Nautile-370M uses a hybrid backbone in which two SeqCond Attention (SCA) layers, a linear-time spectral sequence operator inspired by SeqCondenser, alternate with one transformer layer. This design aims to retain the long-context efficiency and state-tracking benefits of structured sequential models while preserving the expressive token-to-token routing of attention.