TAAM:Inductive Graph-Class Incremental Learning with Task-Aware Adaptive Modulation

2026-02-08

AI Summary

This work addresses key challenges in graph continual learning—namely, the memory overhead and privacy risks associated with replay strategies, the stability-plasticity trade-off, unknown task identities, and data leakage in existing benchmarks—by proposing a Task-Aware Adaptive Modulation (TAAM) framework. TAAM trains and freezes lightweight Neural Synaptic Modulators (NSMs) atop a fixed GNN backbone for each new task, enabling node-level computational flow modulation without replay to effectively mitigate catastrophic forgetting. Additionally, the authors introduce and theoretically justify an Anchored Multi-hop Propagation (AMP) mechanism to handle unknown task IDs and establish a more rigorous inductive evaluation benchmark. Extensive experiments across eight datasets demonstrate that TAAM significantly outperforms state-of-the-art methods in preserving prior knowledge, adapting to new tasks, and operating under unknown task identity scenarios.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Multiagent Systems: Adversarial Agents
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Graph Continual Learning (GCL) aims to solve the challenges of streaming graph data. However, current methods often depend on replay-based strategies, which raise concerns like memory limits and privacy issues, while also struggling to resolve the stability-plasticity dilemma. In this paper, we suggest that lightweight, task-specific modules can effectively guide the reasoning process of a fixed GNN backbone. Based on this idea, we propose Task-Aware Adaptive Modulation (TAAM). The key component of TAAM is its lightweight Neural Synapse Modulators (NSMs). For each new task, a dedicated NSM is trained and then frozen, acting as an"expert module."These modules perform detailed, node-attentive adaptive modulation on the computational flow of a shared GNN backbone. This setup ensures that new knowledge is kept within compact, task-specific modules, naturally preventing catastrophic forgetting without using any data replay. Additionally, to address the important challenge of unknown task IDs in real-world scenarios, we propose and theoretically prove a novel method named Anchored Multi-hop Propagation (AMP). Notably, we find that existing GCL benchmarks have flaws that can cause data leakage and biased evaluations. Therefore, we conduct all experiments in a more rigorous inductive learning scenario. Extensive experiments show that TAAM comprehensively outperforms state-of-the-art methods across eight datasets. Code and Datasets are available at: https://github.com/1iuJT/TAAM_AAMAS2026.

Problem

• Graph Continual Learning
• catastrophic forgetting
• task-agnostic learning
• data replay
• inductive learning

Innovation

• Task-Aware Adaptive Modulation
• Neural Synapse Modulators
• Graph Continual Learning
• Replay-Free Learning
• Inductive Learning