AdaParse: Personalized Fingerprinting for Visual Generative Model Reverse Engineering
- 5 March 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 21 (15566013) , 2682-2697
Abstract
In this paper, we propose a hyperparameter-specialized adaptive fingerprinting framework named AdaParse for model reverse engineering, which aims at predicting hyperparameters of interest in generative models from the given AI-generated images. Existing methods rely on a single coarse model fingerprint that is originally designed for model-level attribution, which makes it difficult to distinguish fine-grained traces corresponding to different hyperparameter configurations in a multitude of generative models. To address this, our AdaParse dynamically responds to instance-level variations by estimating hyperparameter-specific fingerprints via personalizing estimation networks tailored for each input image. Specifically, our approach simultaneously learns two-branch hypernetworks that balance instance-aware and model-agnostic prior knowledge for fingerprint generation. To enable efficient network personalization, we further propose a Broadcasted Fusion module that transforms condensed feature codes into adaptive parameters through factorized weight generation with enhanced representative capacity. Extensive experiments on the large-scale public dataset across 123 generative models demonstrate that our approach outperforms previous state-of-the-art methods. Code available at https://github.com/lizhuoxun/AdaParse/.Keywords
Funding Information
- National Natural Science Foundation of China (62125603)
- National Natural Science Foundation of China (62336004)
- National Natural Science Foundation of China (62321005)
- National Natural Science Foundation of China (62441616)
- National Natural Science Foundation of China (62406172)
- China Postdoctoral Science Foundation (2024M761674)
- China Postdoctoral Science Foundation (2023M741964)
- Postdoctoral Fellowship Program of CPSF (GZC20240841)
This publication has 77 references indexed in Scilit:
- Investigating Style Similarity in Diffusion Models2024
- Which Model Generated This Image? A Model-Agnostic Approach for Origin Attribution2024
- Rethinking Open-World DeepFake Attribution with Multi-perspective Sensory LearningInternational Journal of Computer Vision, 2024
- RepMix: Representation Mixing for Robust Attribution of Synthesized Images2022
- An Information Theoretic Approach for Attention-Driven Face Forgery Detection2022
- Two-Branch Recurrent Network for Isolating Deepfakes in Videos2020
- Thinking in Frequency: Face Forgery Detection by Mining Frequency-Aware Clues2020
- Towards Reverse-Engineering Black-Box Neural Networks2019
- Fighting Fake News: Image Splice Detection via Learned Self-Consistency2018
- BusterNet: Detecting Copy-Move Image Forgery with Source/Target Localization2018