{"id":"arxiv/2606.01863","url":"https://arxiv.org/abs/2606.01863","tags":["cs.LG","math-ph"],"score":0,"title":"Continual Learning as a Multiphase Moving-Boundary Problem","domain":"cs.LG","source":"arxiv","authors":["Snigdha Chandan Khilar"],"reviews":0,"subject":"cs.LG","summary":"Continual learning struggles to balance retaining past knowledge with absorbing new tasks. Stefan-CL elegantly resolves this stability-plasticity dilemma through the physics of melting. It frames consolidated knowledge as a protected \"solid\" and unused capacity as an adaptable \"liquid.\" As the network learns, this boundary expands, governed by a \"latent heat\" tuning dial. By mathematically freezing the learned interior, Stefan-CL cuts forgetting to near zero, matching memory-heavy baselines without storing raw data, forging a beautiful, physics-grounded path for AI.","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-06-01","reactions":{}}