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

arXiv: https://arxiv.org/abs/2410.14102 · HTML: https://mlsanity.com/paper/2410.14102/

## 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.

## Placed in

- LLM safety and security / [Watermarking generated content and models](https://mlsanity.com/safety/watermarking-generated-content-and-models/)

## Builds on (most cited first, 21 of 21)

- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation](https://mlsanity.com/paper/2109.00859/) (2021)
- [Practical Black-Box Attacks against Machine Learning](https://mlsanity.com/paper/1602.02697/) (2016)
- [CodeXGLUE: A Machine Learning Benchmark Dataset for Code Understanding and Generation](https://mlsanity.com/paper/2102.04664/) (2021)
- [Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidates](https://mlsanity.com/paper/1804.10959/) (2018)
- [Stealing Machine Learning Models via Prediction APIs](https://mlsanity.com/paper/1609.02943/) (2016)
- [Detecting Backdoor Attacks on Deep Neural Networks by Activation Clustering](https://mlsanity.com/paper/1811.03728/) (2018)
- [Embedding Watermarks into Deep Neural Networks](https://mlsanity.com/paper/1701.04082/) (2017)
- [code2seq: Generating Sequences from Structured Representations of Code](https://mlsanity.com/paper/1808.01400/) (2018)
- [A Neural Model for Generating Natural Language Summaries of Program Subroutines](https://mlsanity.com/paper/1902.01954/) (2019)
- [Stealthy Backdoor Attack for Code Models](https://mlsanity.com/paper/2301.02496/) (2023)
- [CoProtector: Protect Open-Source Code against Unauthorized Training Usage with Data Poisoning](https://mlsanity.com/paper/2110.12925/) (2021)
- [Digital Watermarking for Deep Neural Networks](https://mlsanity.com/paper/1802.02601/) (2018)
- [CodeMark: Imperceptible Watermarking for Code Datasets against Neural Code Completion Models](https://mlsanity.com/paper/2308.14401/) (2023)
- [PromptCARE: Prompt Copyright Protection by Watermark Injection and Verification](https://mlsanity.com/paper/2308.02816/) (2023)
- [Training-free Lexical Backdoor Attacks on Language Models](https://mlsanity.com/paper/2302.04116/) (2023)
- [Code Structure Guided Transformer for Source Code Summarization](https://mlsanity.com/paper/2104.09340/) (2021)
- [An Extractive-and-Abstractive Framework for Source Code Summarization](https://mlsanity.com/paper/2206.07245/) (2022)
- [On the Difficulty of Defending Contrastive Learning against Backdoor Attacks](https://mlsanity.com/paper/2312.09057/) (2023)
- [ESALE: Enhancing Code-Summary Alignment Learning for Source Code Summarization](https://mlsanity.com/paper/2407.01646/) (2024)
- [Learning Deep Semantic Model for Code Search using CodeSearchNet Corpus](https://mlsanity.com/paper/2201.11313/) (2022)

## Built on by (most cited first, 2 of 2)

- [Towards Generalized and Stealthy Watermarking for Generative Code Models](https://mlsanity.com/paper/2506.20926/) (2025)
- [PuzzleMark: Implicit Jigsaw Learning for Robust Code Dataset Watermarking in Neural Code Completion Models](https://mlsanity.com/paper/2604.27677/) (2026)

## 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.