LSA-PAST: Layer-Specific Adaptation and Prototype Transport for Compositional Incremental Learning

arXiv cs.AIOctober 21, 2026en

arXiv:2610.27203v1 Announce Type: new Abstract: Class-Incremental Learning (CIL) enables models to recognize new classes over time while retaining prior knowledge, but many real-world problems are compositional, where visual concepts arise from combinations of primitives such as states (i.e., red, shiny) and objects (i.e., shoe, car). Standard CIL methods treat each composition as an independent class, ignoring shared primitives and providing no mechanism to stabilize their representations as new compositions arrive. This causes parameter updates to entangle primitives and induce drift that degrades performance on earlier compositions. Compositional Incremental Learning (composition-IL) form

This is a short summary published by AI Global Wire. The full article is owned and hosted by arXiv cs.AI — open it there to read it in full.

• Forskning