IP2-RSNN is introduced, a two-stage framework that allocates intrinsic neuronal plasticity at the task-family level and evaluates its effects during sequential learning, revealing that learning capacity allocation is a principled route towards stable and interpretable sequential learning.
This work proposes Spatially Contiguous Local Attention with Boundary Continuity Pathway (SCLA-BCP), which computes attention within non-overlapping regions of spatially adjacent tokens, while BCP facilitates cross-boundary information exchange through a lightweight convolutional pathway.
Zeqi Zheng, Zizheng Zhu, Yuping Yan et al.· 0 citations
This work introduces a Temporal Feedback Coding (TFC) scheme that leverages feedback at the encoding stage to diversify spike patterns and designs a Global-Local Dynamic LIF (GLD-LIF) neuron that enhances cross-step dependency modeling by integrating local aggregation and global initialization.