analiza aiqbench
Hyperbolic Prototype Routing (HyPro) addresses catastrophic forgetting in class-incremental learning by isolating representations via LoRA experts and utilizing hyperbolic geometry for task discrimination. This rehearsal-free approach demonstrates consistent accuracy improvements on standard benchmarks, offering a viable path toward more robust continual learning systems without requiring stored data.
Hyperbolic Prototype Routing for Rehearsal-Free Class-Incremental Learning
arXiv:2609.39550v1 Announce Type: new Abstract: Class-Incremental Learning (CIL) aims to continually learn new classes while preserving prior knowledge. Parameter-efficient fine-tuning with pre-trained models enables CIL with minimal parameter updates, but existing approaches still suffer from catastrophic forgetting caused by cumulative interference and suboptimal module-sample matching at inference. We propose Hyperbolic Prototype Routing (HyPro), a rehearsal-free framework for continual learning. HyPro allocates a dedicated LoRA-Expert module to each incremental task for isolated representation learning, then projects routing features onto a Poincare ball and performs geodesic nearest-prototype matching for reliable task-level discrimination. Extensive experiments on standard CIL and Few-Shot CIL benchmarks show that HyPro consistently improves average and final-stage accuracy over strong baselines.
Abstract:Class-Incremental Learning (CIL) aims to continually learn new classes while preserving prior knowledge. Parameter-efficient fine-tuning with pre-trained models enables CIL with minimal parameter updates, but existing approaches still suffer from catastrophic forgetting caused by cumulative interference and suboptimal module-sample matching at inference. We propose Hyperbolic Prototype Routing (HyPro), a rehearsal-free framework for continual learning. HyPro allocates a dedicated LoRA-Expert module to each incremental task for isolated representation learning, then projects routing features onto a Poincare ball and performs geodesic nearest-prototype matching for reliable task-level discrimination. Extensive experiments on standard CIL and Few-Shot CIL benchmarks show that HyPro consistently improves average and final-stage accuracy over strong baselines.
Machine Learning (cs.LG)
arXiv:2609.39550 [cs.LG]
(or arXiv:2609.39550v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2609.39550 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
supplementary.pdf
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 201 cuvinte
Subiecte conexe
Distribution-Conditioned Task Routing for Class-Incremental Learning
arXiv · 30 sept. 2026
Dynamic LoRA-Experts and Prototype-Ensemble Matching for Class-Incremental Learning
arXiv · 1 oct. 2026
Dual-Mode Low-Rank Learner with Bridge-Prototype Ensemble for Vision-Language Class-Incremental Learning
arXiv · 30 sept. 2026
OpenRouter: from Seed to Stripe — with OpenRouter’s Alex Atallah & AMP’s Anjney Midha
Latent Space · 25 sept. 2026