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

**Published:** October 21, 2026
**Author:** Kyle Stein, Andrew Arash Mahyari, Guillermo Francia III, Eman El-Sheikh, Jiangpeng He
**Source:** arXiv — cs.AI
**Language:** en-US
**HTML:** https://aiqazaqstan.com/en/articles/lsa-past-layer-specific-adaptation-and-prototype-transport-for-compositi

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…

> From the original source: **arXiv — cs.AI**
> [Read full article at arXiv — cs.AI](https://arxiv.org/abs/2610.27203)

**Tags:** #Papers with Code #arXiv

---

*AI Qazaqstan — non-commercial informational portal about AI in Kazakhstan. Machine translations and summaries are AI-generated.*