Lune

AAAI2025Top-tier venue

Model Lineage Closeness Analysis

Chen Tang, Lan Zhang, Qi Zhao, Xirong Zhuang, Xiang-Yang Li

Abstract

As machine learning model modification techniques are extensively employed to obtain well-performing models at reduced costs, several studies have emerged to determine the presence of a modification relationship (i.e., lineage) between models. However, these methods are not robust to high-impact modification techniques and none of them have addressed the measurement of lineage closeness, which quantifies the degrees of modification. In this work, we visualize the changes in model decision boundaries resulting from different modification techniques and conclude that differences in decision boundaries serve as a precise metric of lineage closeness. Building upon this insight, we propose a modification-type agnostic and task-agnostic method to measure model lineage closeness by calculating mean adversarial distances from data points to decision boundaries and matching rate of data points, with data points selected through an efficient sampling method to reduce computational overhead. Moreover, we propose a novel indirect measurement approach to support lineage closeness measurement for models with different tasks. Finally, comprehensive experiments show that our design achieves an impressive 97% accuracy in lineage determination, and can precisely measure model lineage closeness for different modifications.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Are You Stealing My Model? Sample Correlation for Fingerprinting Deep Neural Networks cites Fingerprinting Deep Neural Networks Globally via Universal Adversarial Perturbations
  • Are You Stealing My Model? Sample Correlation for Fingerprinting Deep Neural Networks cites Copy, Right? A Testing Framework for Copyright Protection of Deep Learning Models
  • Are You Stealing My Model? Sample Correlation for Fingerprinting Deep Neural Networks cites Deep Neural Network Fingerprinting by Conferrable Adversarial Examples
  • Are You Stealing My Model? Sample Correlation for Fingerprinting Deep Neural Networks cites Turning Your Weakness Into a Strength: Watermarking Deep Neural Networks by Backdooring
  • Fingerprinting Deep Neural Networks Globally via Universal Adversarial Perturbations cites Deep Neural Network Fingerprinting by Conferrable Adversarial Examples
  • Fingerprinting Deep Neural Networks Globally via Universal Adversarial Perturbations cites Turning Your Weakness Into a Strength: Watermarking Deep Neural Networks by Backdooring
  • Copy, Right? A Testing Framework for Copyright Protection of Deep Learning Models cites Deep Neural Network Fingerprinting by Conferrable Adversarial Examples
  • Copy, Right? A Testing Framework for Copyright Protection of Deep Learning Models cites Turning Your Weakness Into a Strength: Watermarking Deep Neural Networks by Backdooring
  • Deep Neural Network Fingerprinting by Conferrable Adversarial Examples cites Turning Your Weakness Into a Strength: Watermarking Deep Neural Networks by Backdooring
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on8

  1. Turning Your Weakness Into a Strength: Watermarking Deep Neural Networks by BackdooringYossi Adi, Carsten Baum, Moustapha Cissé, Benny Pinkas et al.USENIX Security 2018 · 832 citations
  2. Deep Neural Network Fingerprinting by Conferrable Adversarial ExamplesNils Lukas, Yuxuan Zhang, Florian KerschbaumICLR 2021 · 182 citations
  3. Copy, Right? A Testing Framework for Copyright Protection of Deep Learning ModelsJialuo Chen, Jingyi Wang, Tinglan Peng, Youcheng Sun et al.S&P 2022 · 94 citations
  4. Fingerprinting Deep Neural Networks Globally via Universal Adversarial PerturbationsZirui Peng, Shaofeng Li, Guoxing Chen, Cheng Zhang et al.CVPR 2022 · 66 citations
  5. Are You Stealing My Model? Sample Correlation for Fingerprinting Deep Neural NetworksJiyang Guan, Jian Liang, Ran HeNeurIPS 2022 · 57 citations

Related papers

  1. Neural LineageRunpeng Yu, Xinchao WangCVPR 2024
  2. An In-Depth Study on Deep Learning Model CloningBin Hu, Xiancong Pan, Dongjin Yu, Tianyi HuICML 2026
  3. Hold me tight! Influence of discriminative features on deep network boundariesGuillermo Ortiz-Jiménez, Apostolos Modas, Seyed-Mohsen Moosavi-Dezfooli, Pascal FrossardNeurIPS 2020 · 53 citations
  4. Leveraging Adversarial Examples to Quantify Membership Information LeakageGanesh Del Grosso, Hamid Jalalzai, Georg Pichler, Catuscia Palamidessi et al.CVPR 2022 · 2 citations
  5. ModelDiff: testing-based DNN similarity comparison for model reuse detectionYuanchun Li, Ziqi Zhang, Bingyan Liu, Ziyue Yang et al.ISSTA 2021 · 44 citations
Dusk over the sea between two cliffs drawn in fine vertical lines