Hyperbolic Prototype Routing for Rehearsal-Free Class-Incremental Learning

HongWei Zhao; Rui Liu; Yong Chen

cs.LGSep 30, 2026

Hyperbolic Prototype Routing for Rehearsal-Free Class-Incremental Learning

Authors: HongWei Zhao, Rui Liu, Yong Chen

Organizations: Beihang University · Beijing University of Posts and Telecommunications

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.

Figures & tables

Fig. 1 : Illustration of HyPro. In the $t$ -th incremental task, a new LoRA-Expert $\mathbf{E}_{t}$ (with parameters $\mathbf{A}_{t}$ and $\mathbf{B}_{t}$ ) is trained to capture task-specific features. Domain-specific features from the router $\mathbf{E}^{router}$ are projected into the hyperbolic space (Poincaré ball). During inference, the nearest prototype guides LoRA-Experts selection for each input sample.

Method	CIFAR100 ( $T$ =10)		CUB200 ( $T$ =10)
Method	$\mathcal{A}_{L}$	$\bar{\mathcal{A}}$	$\mathcal{A}_{L}$	$\bar{\mathcal{A}}$
Full Fine-Tuning	$66.26$	$76.94$	$55.29$	$70.30$
SimpleCIL [ 7 ]	$81.27$	$87.13$	$82.28$	$91.85$
L2P [ 5 ]	$84.82$	$89.78$	$71.98$	$81.80$
CODA-Prompt [ 6 ]	$86.69$	$91.31$	$75.45$	$84.65$
InfLoRA [ 8 ]	$86.43$	$91.80$	$70.07$	$81.71$

TABLE I : Performance comparison of selected CIL methods, all built on the same pre-trained backbone ( ViT-B/16-IN21K ).

Method	CUB200 ( $T$ =11)			CIFAR100 ( $T$ =9)
Method	$\mathcal{A}_{\text{Base}}$	$\mathcal{A}_{L}$	$\bar{\mathcal{A}}$	$\mathcal{A}_{\text{Base}}$	$\mathcal{A}_{L}$	$\bar{\mathcal{A}}$
L2P [ 5 ]	$91.50$	$50.04$	$66.70$	$93.43$	$55.75$	$71.81$
CODA-Prompt [ 6 ]	$91.50$	$53.65$	$69.30$	$94.05$	$57.10$	$73.11$
InfLoRA [ 8 ]	$92.45$	$45.18$	$66.27$	$\bf{94.92}$	$57.41$	$74.28$
SD-LoRA [ 9 ]	$91.92$	$56.28$	$70.87$	$94.60$	$73.51$	$78.42$
CPE-CLIP [ 32 ]	$80.21$	$63.32$	$69.37$	$88.32$	$79.99$	$83.38$

TABLE II : Performance comparison of selected FSCIL methods, all built on the same pre-trained backbone ( ViT-B/16-IN21K ).

Ablated Components	ImageNet-R ( $T$ =5)		CIFAR100 ( $T$ =9)
Ablated Components	$\mathcal{A}_{L}$	$\bar{\mathcal{A}}$	$\mathcal{A}_{L}$	$\bar{\mathcal{A}}$
w/o LoRA Dynamically	61.17 / 72.37	74.31 / 80.14	73.03 / 79.73	80.39 / 86.59
w/o HPR	69.53 / 73.13	77.75 / 80.42	81.13 / 84.09	86.97 / 88.67
HyPro-MLP / QV	77.00 / 78.10	82.44 / 83.05	88.96 / 88.08	91.16 / 90.54

TABLE III : Ablation studies on CIL and FSCIL tasks. The first dataset corresponds to CIL, and the second to FSCIL. For each metric, the left/right values represent performance with MLP-LoRA and QV-LoRA fine-tuning, respectively. HPR: Hyperbolic Prototype Routing.

Method	CIFAR100 ( $T$ =10)	CUB200 ( $T$ =10)	ImageNet-R ( $T$ =5)
KNN	86.80	90.40	75.24
Prototype	89.60	91.10	77.04
HyPro-MLP	93.71	93.40	88.48
HyPro-QV	94.13	93.32	88.61

TABLE IV : Router average accuracy comparison of different module-sample matching strategies on CIL tasks. All methods are based on the same pre-trained backbone ( ViT-B/16-IN21K ).

Explore similar work

Dynamic LoRA-Experts and Prototype-Ensemble Matching for Class-Incremental Learning
Sep 30, 2026Hongwei Zhao, Rui Liu, Yansong LiuClass-Incremental Learning Layer-Selective Rehearsal Strategy
School of Computer Science and Engineering, Beihang University
Miles: Metric Learning with Expandable Subspace for Pre-Trained Model-Based Class-Incremental Learning
Jul 20, 2026Kai Jiang, Zisong Lin, Hongyuan Zhang +2Class-Incremental Learning Meta-Learning
National Key Laboratory of Radar Signal Processing, Xidian University, Xi’an 710071, China · The University of Hong Kong, Hong Kong SAR, China · Institute of Artificial Intelligence (TeleAI) of China Telecom, China
HydraCIL: Decoupled Class-Incremental Learning through Prototype-Guided Multi-Head Classifiers
Jun 8, 2026Daniel Vila-Cruz, Laura Morán-Fernández, Verónica Bolón-CanedoClass-Incremental Learning Continual Learning
Universidade da Coruña CITIC A Coruña, Spain

Sep 30, 2026, cs.LGJ/K move · Enter open · S save

Dynamic LoRA-Experts and Prototype-Ensemble Matching for Class-Incremental Learning

Hongwei Zhao, Rui Liu, Yansong Liu

School of Computer Science and Engineering, Beihang University

Class-Incremental Learning Layer-Selective Rehearsal Strategy

Open paper AskPDF arXiv

Class-Incremental Learning (CIL) aims to continuously learn new classes without forgetting previously acquired knowledge. Parameter-efficient fine-tuning with pre-trained models reduces parameter overhead but can suffer from cumulative interference and suboptimal alignment between inference samples and specialized modules. We propose Dynamic LoRA-Experts and Prototype-Ensemble Matching (DLEPEM), a two-stage rehearsal-free framework. DLEPEM allocates a task-specific LoRA-Expert for each incremental task to reduce cross-task interference, then combines frozen pre-trained-model prototypes with task-adaptive LoRA-Expert prototypes for reliable task-level discrimination. Experiments on standard CIL and Few-Shot CIL benchmarks demonstrate strong performance under the evaluated protocols.

Abstract

Explore similar work

Dynamic LoRA-Experts and Prototype-Ensemble Matching for Class-Incremental Learning

Miles: Metric Learning with Expandable Subspace for Pre-Trained Model-Based Class-Incremental Learning

HydraCIL: Decoupled Class-Incremental Learning through Prototype-Guided Multi-Head Classifiers