Q. Wang, “DualCP: Rehearsal-Free Domain-Incremental Learning via Dual-Level Concept Prototype”, AAAI, vol. 39, no. 20, pp. 21198–21206, Apr. 2025.