LSA-PAST: Layer-Specific Adaptation and Prototype Transport for Compositional Incremental Learning
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,…
Read the full story at arXiv cs.AI ↗
Timeline · 1 report
- 2026-10-21 04:00 · arXiv cs.AI
LSA-PAST: Layer-Specific Adaptation and Prototype Transport for Compositional Incremental Learning