RESEARCH

NeuMoSync: End-to-End Neuromodulatory Control for Plasticity and Adaptability in Continual Learning

ArXiv cs.AI · Thu, 06 Aug 2026 04:00:00 GMT

arXiv:2608.04358v1 Announce Type: new Abstract: Continual learning (CL) requires models to learn tasks sequentially, yet deep neural networks often suffer from plasticity loss and poor knowledge transfer, which can impede their long-term adaptability. Drawing high-level inspirati

Read original source Discuss with SiiMON