Understanding Imbalanced Forgetting in Rehearsal-Based Class-Incremental Learning

ArXi:2605.14785v1 Announce Type: new Neural networks suffer from catastrophic forgetting in class-incremental learning (CIL) settings. Rehearsal$\unicode{x2013}$replaying a subset of past samples$\unicode{x2013}$is a well-established mitigation strategy. However, recent results suggest that, despite balanced rehearsal allocation, some classes are forgotten substantially than others. Despite its relevance, this imbalanced forgetting phenomenon remains underexplored.