mlsanityData to 14 Oct 2026

Towards the Resistance of Neural Network Watermarking to Fine-tuning

arXiv 2505.01007 ↗First posted 2 May 2025cs.LG cs.AI cs.CLcited by 2

Not on the map. Why?

Abstract (arXiv)

This paper proves a new watermarking method to embed the ownership information into a deep neural network (DNN), which is robust to fine-tuning. Specifically, we prove that when the input feature of a convolutional layer only contains low-frequency components, specific frequency components of the convolutional filter will not be changed by gradient descent during the fine-tuning process, where we propose a revised Fourier transform to extract frequency components from the convolutional filter. Additionally, we also prove that these frequency components are equivariant to weight scaling and weight permutations. In this way, we design a watermark module to encode the watermark information to specific frequency components in a convolutional filter. Preliminary experiments demonstrate the effectiveness of our method.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 13 of its 24 references that are in the map. First posted on arXiv 2 May 2025.

Builds on

13 of 24 references.

References · 13

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2019Similarity of Neural Network Representations Revisitedcited by 1,755
  3. 2023A Watermark for Large Language Modelscited by 725
  4. 2018Turning Your Weakness Into a Strength: Watermarking Deep Neural Networks by Backdooringcited by 357
  5. 2017Embedding Watermarks into Deep Neural Networkscited by 248
  6. 2019Deep Neural Network Fingerprinting by Conferrable Adversarial Examplescited by 93
  7. 2024Instructional Fingerprinting of Large Language Modelscited by 68
  8. 2022Certified Neural Network Watermarks with Randomized Smoothingcited by 41
  9. 2024REEF: Representation Encoding Fingerprints for Large Language Modelscited by 36
  10. 2023HuRef: HUman-REadable Fingerprint for Large Language Modelscited by 34
  11. 2020Neural Network Laundering: Removing Black-Box Backdoor Watermarks from Deep Neural Networkscited by 27
  12. 2023Functional Invariants to Watermark Large Transformerscited by 15
1 more
  1. 2022Defects of Convolutional Decoder Networks in Frequency Representationcited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 1.

2 papers.

  1. 2025LLM DNA: Tracing Model Evolution via Functional Representationscited by 15
  2. 2026AttnDiff: Attention-based Differential Fingerprinting for Large Language Modelscited by 4

Similar, not cited

Close in wording, no citation either way

  1. 2022Free Fine-tuning: A Plug-and-Play Watermarking Scheme for Deep Neural Networks
  2. 2023On Function-Coupled Watermarks for Deep Neural Networks
  3. 2021A survey of deep neural network watermarking techniques
  4. 2024FreeMark: A Non-Invasive White-Box Watermarking for Deep Neural Networks
  5. 2022Neural network fragile watermarking with no model performance degradation
  6. 2025Hashed Watermark as a Filter: Defeating Forging and Overwriting Attacks in Weight-based Neural Network Watermarking
  7. 2025Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation
  8. 2022Deep Fidelity in DNN Watermarking: A Study of Backdoor Watermarking for Classification Models