mlsanityData to 14 Oct 2026

Beyond Dataset Watermarking: Model-Level Copyright Protection for Code Summarization Models

arXiv 2410.14102 ↗First posted 18 Oct 2024cs.CR cs.CYcited by 2

On the map: LLM safety and security → Watermarking generated content and models

Abstract (arXiv)

Code Summarization Model (CSM) has been widely used in code production, such as online and web programming for PHP and Javascript. CSMs are essential tools in code production, enhancing software development efficiency and driving innovation in automated code analysis. However, CSMs face risks of exploitation by unauthorized users, particularly in an online environment where CSMs can be easily shared and disseminated. To address these risks, digital watermarks offer a promising solution by embedding imperceptible signatures within the models to assert copyright ownership and track unauthorized usage. Traditional watermarking for CSM copyright protection faces two main challenges: 1) dataset watermarking methods require separate design of triggers and watermark features based on the characteristics of different programming languages, which not only increases the computation complexity but also leads to a lack of generalization, 2) existing watermarks based on code style transformation are easily identifiable by automated detection, demonstrating poor concealment. To tackle these issues, we propose ModMark , a novel model-level digital watermark embedding method. Specifically, by fine-tuning the tokenizer, ModMark achieves cross-language generalization while reducing the complexity of watermark design. Moreover, we employ code noise injection techniques to effectively prevent trigger detection. Experimental results show that our method can achieve 100% watermark verification rate across various programming languages' CSMs, and the concealment and effectiveness of ModMark can also be guaranteed.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Watermarking generated content and models, one of 12 directions in LLM safety and security. Builds on 21 of its 49 references that are in the map. First posted on arXiv 18 Oct 2024.

Builds on

21 of 49 references.

Closest ancestors · 1

Same direction, newest first

  1. 2023CodeMark: Imperceptible Watermarking for Code Datasets against Neural Code Completion Models28 Aug

Other references · 20

  1. 2016SQuAD: 100,000+ Questions for Machine Comprehension of Textcited by 4,148
  2. 2021CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generationcited by 906
  3. 2016Practical Black-Box Attacks against Machine Learningcited by 903
  4. 2021CodeXGLUE: A Machine Learning Benchmark Dataset for Code Understanding and Generationcited by 744
  5. 2018Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidatescited by 639
  6. 2016Stealing Machine Learning Models via Prediction APIscited by 619
  7. 2018Detecting Backdoor Attacks on Deep Neural Networks by Activation Clusteringcited by 456
  8. 2017Embedding Watermarks into Deep Neural Networkscited by 248
  9. 2018code2seq: Generating Sequences from Structured Representations of Codecited by 179
  10. 2019A Neural Model for Generating Natural Language Summaries of Program Subroutinescited by 77
  11. 2023Stealthy Backdoor Attack for Code Modelscited by 49
  12. 2021CoProtector: Protect Open-Source Code against Unauthorized Training Usage with Data Poisoningcited by 47
8 more
  1. 2018Digital Watermarking for Deep Neural Networkscited by 39
  2. 2023PromptCARE: Prompt Copyright Protection by Watermark Injection and Verificationcited by 32
  3. 2023Training-free Lexical Backdoor Attacks on Language Modelscited by 31
  4. 2021Code Structure Guided Transformer for Source Code Summarizationcited by 30
  5. 2022An Extractive-and-Abstractive Framework for Source Code Summarizationcited by 18
  6. 2023On the Difficulty of Defending Contrastive Learning against Backdoor Attackscited by 9
  7. 2024ESALE: Enhancing Code-Summary Alignment Learning for Source Code Summarizationcited by 8
  8. 2022Learning Deep Semantic Model for Code Search using CodeSearchNet Corpuscited by 1

Built on by

Cited by year (2026 to 14 Oct)

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

2 papers.

  1. 2025Towards Generalized and Stealthy Watermarking for Generative Code Modelscited by 0
  2. 2026PuzzleMark: Implicit Jigsaw Learning for Robust Code Dataset Watermarking in Neural Code Completion Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Is The Watermarking Of LLM-Generated Code Robust?
  2. 2026DuCodeMark: Dual-Purpose Code Dataset Watermarking via Style-Aware Watermark-Poison Design ↗
  3. 2025Copyright Protection for Large Language Models: A Survey of Methods, Challenges, and Trends
  4. 2025DERMARK: A Dynamic, Efficient and Robust Multi-bit Watermark for Large Language Models
  5. 2026Preserving Structure, Strengthening Signals: A Grammar-Driven Approach to Optimize the Trade-off Between Quality and Detectability in Code Watermarking
  6. 2025Leave No TRACE: Black-box Detection of Copyrighted Dataset Usage in Large Language Models via Watermarking
  7. 2025CLASP: Training-Free LLM-Assisted Source Code Watermarking via Semantic-Preserving Transformations
  8. 2024Watermarking Large Language Models and the Generated Content: Opportunities and Challenges