DRIFT: A Benchmark for Task-Free Continual Graph Learning with Continuous Distribution Shifts

TL;DR

This paper introduces DRIFT, a benchmark for task-free continual graph learning that models continuous distribution shifts, revealing the limitations of existing methods under realistic non-stationary conditions.

Contribution

It proposes a unified task-free formulation for continual graph learning and provides a benchmark spanning various transition dynamics, highlighting the need for methods that handle continuous distribution drift.

Findings

Existing methods rely on task boundaries and perform poorly under continuous drift.

The DRIFT benchmark covers a spectrum from task switches to smooth distributional changes.

Current approaches struggle with realistic non-stationary graph streams.

Abstract

Continual graph learning (CGL) aims to learn from dynamically evolving graphs while mitigating catastrophic forgetting. Existing CGL approaches typically adopt a task-based formulation, where the data stream is partitioned into a sequence of discrete tasks with pre-defined boundaries. However, such assumptions rarely hold in real-world environments, where data distributions evolve continuously and task identity is often unavailable. To better reflect realistic non-stationary environments, we revisit continual graph learning from a task-free perspective. We propose a unified formulation that models the data stream as a time-varying mixture of latent task distributions, enabling continuous modeling of distribution drift. Based on this formulation, we construct \emph{DRIFT}, a benchmark that spans a spectrum of transition dynamics ranging from hard task switches to smooth distributional drift through a Gaussian parameterization. We evaluate representative continual learning methods under this task-free setting and observe substantial performance degradation compared to traditional task-based protocols. Our findings indicate that many existing approaches implicitly rely on task boundary information and struggle under realistic task-free graph streams. This work highlights the importance of studying continual graph learning under realistic non-stationary conditions and provides a benchmark for future research in this direction. Our code is available at https://github.com/UConn-DSIS/DRIFT.