Socialized Division and Collaboration: Rethinking Class-Incremental Learning under Optimization Conflicts

Xinjie Yao; Zhihe Fan; Yunqi Zhu; Jiaqi Zhou; Dengyu Zhao; Zhoupeng Guo; Yan Fan; Guosong Jiang

cs.AIAug 21, 2026

Socialized Division and Collaboration: Rethinking Class-Incremental Learning under Optimization Conflicts

Authors: Xinjie Yao, Zhihe Fan, Yunqi Zhu, Jiaqi Zhou, Dengyu Zhao, Zhoupeng Guo, Yan Fan, Guosong Jiang,

Organizations: Faculty of Information Engineering and Automation, Kunming University of Science and Technology · School of Sports Training, Tianjin University of Sport · School of Computer Science and Engineering, University of New South Wales · School of Artificial Intelligence, Tianjin University ·

Abstract

Class-incremental learning is commonly instantiated as a single-model paradigm, where a unified model sequentially adapts to an unbounded stream of sessions. While effective under mild distributional shifts, this formulation becomes strained when successive sessions induce incompatible optimization directions, leading to destructive interference and catastrophic forgetting. We argue that such forgetting reflects a structural limitation of enforcing heterogeneous learning dynamics within a single parameter space. Motivated by social solidarity theory, we propose Socialized Division and Collaboration (SDC) as a reformulation of continual learning that decomposes session learning across specialized models in response to optimization conflicts, while enabling coordinated collaboration. To support this formulation with a principled allocation mechanism, we introduce an energy-based session-model compatibility criterion grounded in Helmholtz free energy, which guides adaptive session allocation and model evolution under conflicting objectives. This framework integrates session assignment, model evolution, and collaborative inference into a unified pipeline, offering an alternative to monolithic continual learning formulations and highlighting a broader design principle for learning under persistent optimization conflicts.

Figures & tables

Figure 1: Social solidarity in human society versus socialized solidarity in machine society.

Figure 2: Overview of the proposed Socialized Division and Collaboration (SDC) framework. SDC consists of a Helmholtz free energy selector for session division, a model evolution module for specialized learning, and a collaboration module for structured integration, jointly enabling division-aware and collaboration-aware continual learning.

Method	ImageNet-R		CIFAR-100		CUB-200
Method	ACC (↑)	AAA (↑)	ACC (↑)	AAA (↑)	ACC (↑)	AAA (↑)
Fine-Tuning ( Shuttleworth et al. 2025 )	60.57	72.31	69.49	80.35	51.43	69.74
L2P ( Wang et al. 2022b )	71.26	76.13	83.18	87.69	65.18	76.12
DualPrompt ( Wang et al. 2022a )	68.22	73.81	81.48	86.41	68.00	79.40
CODA-Prompt ( Smith et al. 2023 )	74.05	78.14	86.31	90.67	71.92	78.76
InfLoRA ( Liang and Li 2024 )	74.75	80.67	86.75	91.72	70.82	81.39

Table 1: Performance of different methods on ImageNet-R, CIFAR-100 and CUB-200. The 1 st and 2 nd best results are highlighted in bold and underlined , respectively.

Method	ImageNet-R		CIFAR-100		CUB-200
Method	ACC ( $\uparrow$ )	AAA ( $\uparrow$ )	ACC ( $\uparrow$ )	AAA ( $\uparrow$ )	ACC ( $\uparrow$ )	AAA ( $\uparrow$ )
SDC-M2	83.99	87.02	93.36	95.87	83.94	90.26
SDC-M3	85.96	88.61	95.08	96.58	84.93	90.88
SDC-M4	85.85	88.29	96.13	97.40	84.80	89.50

Table 2: Effect of the number of models on SDC performance. The 1 st and 2 nd best results are highlighted in bold and underlined , respectively.

Figure 3: Detailed accuracy of each session on ImageNet-R.

Method	HFES	DME	ACC ( $\uparrow$ )		AAA ( $\uparrow$ )
SDC	$\times$	$\times$	70.62		76.32
	$\times$	$\checkmark$	79.86		84.62
	$\checkmark$	$\times$	82.15		86.15
	$\checkmark$	$\checkmark$	85.96	(+3.81)	88.61	(+2.46)

Table 3: Ablation study on ImageNet-R. The 1 st and 2 nd best results are highlighted in bold and underlined , respectively.

Panel of Figure 4 — Figure 4: Comparing the gradient updates of different methods.

Method	GFLOPs	Parameters	Features
		Learnable	Stored
		(M)	(M)
L2P ( Wang et al. 2022b )	70.14	0.48	0
DualPrompt ( Wang et al. 2022a )	70.26	0.06	0
CODA-Prompt ( Smith et al. 2023 )	70.61	0.38	0
HiDe-Prompt ( Wang et al. 2023a )	70.36	0.08	0.10

Table 4: Comparison of computational cost, learnable parameters, and memory overhead across different methods.

Figure 5: Data-driven semantic correlation heatmaps of class centroids across CIFAR-100, CUB-200, and ImageNet-R. The distinct block-diagonal patterns (red blocks) naturally emerge under Ward’s clustering and Two-Slope normalization, demonstrating that the underlying data inherently partitions into three major semantic groups. This inherent distribution strictly dictates our choice of $K=3$ .

Method	ACC ( $\uparrow$ )	AAA ( $\uparrow$ )
Fine-Tuning	64.92	75.57
L2P	73.04	76.94
DualPrompt	69.99	72.24
CODA-Prompt	76.63	80.30
HiDe-Prompt	74.77	78.15
InfLoRA	76.95	81.81

Table 5: Performance of different methods on the ImageNet-R (number of session=5). The 1 st and 2 nd best results are highlighted in bold and underlined , respectively.

Method	ACC ( $\uparrow$ )	AAA ( $\uparrow$ )
Fine-Tuning	49.95	65.32
L2P	68.97	74.16
DualPrompt	65.23	71.30
CODA-Prompt	69.38	73.95
HiDe-Prompt	73.59	77.93
InfLoRA	69.89	76.68

Table 6: Performance of different methods on the ImageNet-R (number of session=20). The 1 st and 2 nd best results are highlighted in bold and underlined , respectively.

Method	ImageNet-R		CIFAR-100		CUB-200
	ACC( $\uparrow$ )	AAA( $\uparrow$ )	ACC( $\uparrow$ )	AAA( $\uparrow$ )	ACC( $\uparrow$ )	AAA( $\uparrow$ )
SDC-M2	83.99	87.02	93.36	95.87	83.94	90.26
SDC-M3	85.96	88.61	95.08	96.58	84.93	90.88
SDC-M4	85.85	88.29	96.13	97.40	84.80	89.50

Table 7: Effect of the number of models on SDC performance.

Division strategy	ImageNet-R		CIFAR-100		CUB-200
Division strategy	ACC ( $\uparrow$ )	AAA ( $\uparrow$ )	ACC ( $\uparrow$ )	AAA ( $\uparrow$ )	ACC ( $\uparrow$ )	AAA ( $\uparrow$ )
Uniform random division	84.58	87.04	92.34	94.75	82.79	88.15
Imbalanced division	83.06	86.12	91.64	94.72	82.13	87.83
Division-aware collaboration	85.96	88.61	95.08	96.58	84.93	90.88

Table 8: Effect of division strategies in SDC on multiple datasets. The 1 st and 2 nd best results are highlighted in bold and underlined , respectively.

Method	HFES	DME	ImageNet-R		CIFAR-100		CUB-200
Method	HFES	DME	ACC	AAA	ACC	AAA	ACC	AAA
SDC	$\times$	$\times$	70.62	76.32	89.54	93.67	75.64	83.19
	$\times$	$\checkmark$	79.86	84.62	89.74	93.68	78.09	84.43
	$\checkmark$	$\times$	82.15	86.15	94.68	95.98	83.97	90.08
	$\checkmark$	$\checkmark$	85.96	88.61	95.08	96.58	84.93	90.88

Table 9: Ablation study of HFES and DME modules across datasets. The 1 st and 2 nd best results are highlighted in bold and underlined , respectively.

Class-Incremental Learning

Class-incremental learning studies how models can add new categories over time while retaining the ability to recognize previously learned classes.

See all 162 papers

Decentralized collaborative continual learning: A multi-objective minimization-based technique
Oct 7, 2026Yara Zgheib, Marc Antonini, Roula NassifDecentralized Learning Replay-Based Continual Learning
Universit´e Cˆote d’Azur, I3S Laboratory, CNRS, France
Free-Flow Class-Incremental Learning: Towards Robust CIL under Variable Class Arrivals
Apr 3, 2026Zhiming Xu, Baile Xu, Jian Zhao +2Class-Incremental Learning
National Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China · School of Artificial Intelligence, Nanjing University, Nanjing, China · School of Electronic Science and Engineering, Nanjing University, China +1
Semantic-Guided Dynamic Sparsification for Pre-Trained Model-based Class-Incremental Learning
Jan 29, 2026Ruiqi Liu, Boyu Diao, Zijia An +3Class-Incremental Learning Activation Sparsity
Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China · University of Chinese Academy of Sciences, Beijing, China

Oct 7, 2026J/K move · Enter open · S save

Decentralized collaborative continual learning: A multi-objective minimization-based technique

Yara Zgheib, Marc Antonini, Roula Nassif

Universit´e Cˆote d’Azur, I3S Laboratory, CNRS, France

Decentralized Learning Replay-Based Continual Learning

Open paper AskPDF arXiv

In this work, we formulate decentralized continual learning within a multi-objective optimization framework. For a given inference task t (corresponding to a common minimizer shared by the cost functions of all agents), agents collecting data in a distributed and streamed manner are only allowed to perform local computations and to exchange information with neighboring agents over the underlying communication graph. As tasks evolve sequentially over time, agents must adapt to newly arriving tasks while retaining knowledge acquired from previously learned ones. This requirement leads to the wellknown stability plasticity dilemma, where stability refers to the ability to retain previous knowledge, while plasticity refers to the ability to learn and adapt to new tasks. To address the stability challenge, agents store subsets of samples from past tasks in local memory buffers. Then, through an appropriate multiobjective formulation, the stored information is incorporated into the learning process so that parameter updates account jointly for the current task and previously learned tasks. The proposed decentralized continual learning approach is analyzed in the mean square error sense under general assumptions on the individual cost functions and gradient noise processes. The analysis reveals that cooperation among agents improves the performance of continual learning. In particular, by exchanging information with neighboring agents, decentralized collaborative learning can exploit the diversity of locally observed data and memory buffers to improve the network average mean-square deviation (MSD) across tasks. Finally, simulations illustrate the theoretical findings and the effectiveness of the method in reducing forgetting and improving the average MSD across tasks.

Abstract

Class-Incremental Learning

Explore similar work

Decentralized collaborative continual learning: A multi-objective minimization-based technique

Free-Flow Class-Incremental Learning: Towards Robust CIL under Variable Class Arrivals

Semantic-Guided Dynamic Sparsification for Pre-Trained Model-based Class-Incremental Learning