HAD: Heterogeneity-Aware Distillation for Lifelong Heterogeneous Learning

ArXi:2603.26192v1 Announce Type: new Lifelong learning aims to preserve knowledge acquired from previous tasks while incorporating knowledge from a sequence of new tasks. However, most prior work explores only streams of homogeneous tasks (\textit{e.g.}, only classification tasks) and neglects the scenario of learning across heterogeneous tasks that possess different structures of outputs. In this work, we formalize this broader setting as lifelong heterogeneous learning