IncreFA: Breaking the Static Wall of Generative Model Attribution

TL;DR

IncreFA introduces a novel incremental learning framework for AI image attribution that adapts continuously to emerging generative models by leveraging hierarchical constraints and a latent memory bank.

Contribution

It redefines attribution as an incremental learning problem, integrating hierarchical architectural constraints and a latent memory bank for continual adaptation and open-set detection.

Findings

Achieves state-of-the-art attribution accuracy on IABench.

Detects 98.93% of unseen models in open-set scenarios.

Effectively adapts to 28 models released between 2022 and 2025.

Abstract

As AI generative models evolve at unprecedented speed, image attribution has become a moving target. New diffusion, adversarial and autoregressive generators appear almost monthly, making existing watermark, classifier and inversion methods obsolete upon release. The core problem lies not in model recognition, but in the inability to adapt attribution itself. We introduce IncreFA, a framework that redefines attribution as a structured incremental learning problem, allowing the system to learn continuously as new generative models emerge. IncreFA departs from conventional incremental learning by exploiting the hierarchical relationships among generative architectures and coupling them with continual adaptation. It integrates two mutually reinforcing mechanisms: (1) Hierarchical Constraints, which encode architectural hierarchies through learnable orthogonal priors to disentangle family-level invariants from model-specific idiosyncrasies; and (2) a Latent Memory Bank, which replays compact latent exemplars and mixes them to generate pseudo-unseen samples, stabilising representation drift and enhancing open-set awareness. On the newly constructed Incremental Attribution Benchmark (IABench) covering 28 generative models released between 2022 and 2025, IncreFA achieves state-of-the-art attribution accuracy and 98.93% unseen detection under a temporally ordered open-set protocol. Code will be available at https://github.com/Ant0ny44/IncreFA.