Efficient and Universal Watermarking for LLM-Generated Code Detection

• URL: http://arxiv.org/abs/2402.07518v4
• Date: Fri, 01 Aug 2025 01:17:22 GMT
• Title: Efficient and Universal Watermarking for LLM-Generated Code Detection
• Authors: Boquan Li, Zirui Fu, Mengdi Zhang, Peixin Zhang, Jun Sun, Xingmei Wang,
• Abstract summary: Large language models (LLMs) have significantly enhanced the usability of AI-generated code.<n>For accountability, it is imperative to detect whether a piece of code is AI-generated.<n>We propose a plug-and-play watermarking approach for AI-generated code detection, named ACW.
• Score: 5.782554045290121
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Large language models (LLMs) have significantly enhanced the usability of AI-generated code, providing effective assistance to programmers. This advancement also raises ethical and legal concerns, such as academic dishonesty or the generation of malicious code. For accountability, it is imperative to detect whether a piece of code is AI-generated. Watermarking is broadly considered a promising solution and has been successfully applied to identify LLM-generated text. However, existing efforts on code are far from ideal, suffering from limited universality and excessive time and memory consumption. In this work, we propose a plug-and-play watermarking approach for AI-generated code detection, named ACW (AI Code Watermarking). ACW is training-free and works by selectively applying a set of carefully-designed, semantic-preserving and idempotent code transformations to LLM code outputs. The presence or absence of the transformations serves as implicit watermarks, enabling the detection of AI-generated code. Our experimental results show that ACW effectively detects AI-generated code, preserves code utility, and is resilient against code optimizations. Especially, ACW is efficient and is universal across different LLMs, addressing the limitations of existing approaches.

Related papers

• Detecting LLM-generated Code with Subtle Modification by Adversarial Training [4.814313782484443]
  We propose an enhanced version of CodeGPTSensor, which employs adversarial training to improve robustness against input perturbations.<n> Experimental results on the HMCorp dataset demonstrate that CodeGPTSensor+ significantly improves detection accuracy on the adversarial test set.
  arXiv  Detail & Related papers  (2025-07-17T13:38:16Z)
• Disappearing Ink: Obfuscation Breaks N-gram Code Watermarks in Theory and Practice [23.788321123219244]
  Distinguishing AI-generated code from human-written code is crucial for authorship attribution, content tracking, and misuse detection.<n>N-gram-based watermarking schemes have emerged as prominent, which inject secret watermarks to be detected during the generation.<n>Most claims rely solely on defenses against simple code transformations or code optimizations as a simulation of attack, creating a questionable sense of robustness.
  arXiv  Detail & Related papers  (2025-07-07T22:18:19Z)
• In-Context Watermarks for Large Language Models [71.29952527565749]
  In-Context Watermarking (ICW) embeds watermarks into generated text solely through prompt engineering.<n>We investigate four ICW strategies at different levels of granularity, each paired with a tailored detection method.<n>Our experiments validate the feasibility of ICW as a model-agnostic, practical watermarking approach.
  arXiv  Detail & Related papers  (2025-05-22T17:24:51Z)
• On-Device Watermarking: A Socio-Technical Imperative For Authenticity In The Age of Generative AI [0.0]
  We argue that we are adopting the wrong approach, and should instead focus on watermarking via cryptographic signatures. For audio-visual content, in particular, all real content is grounded in the physical world and captured via hardware sensors.
  arXiv  Detail & Related papers  (2025-04-15T20:36:52Z)
• SoK: Watermarking for AI-Generated Content [112.9218881276487]
  Watermarking schemes embed hidden signals within AI-generated content to enable reliable detection. Watermarks can play a crucial role in enhancing AI safety and trustworthiness by combating misinformation and deception. This work aims to guide researchers in advancing watermarking methods and applications, and support policymakers in addressing the broader implications of GenAI.
  arXiv  Detail & Related papers  (2024-11-27T16:22:33Z)
• SoK: On the Role and Future of AIGC Watermarking in the Era of Gen-AI [24.187726079290357]
  AIGC watermarks offer an effective solution to mitigate malicious activities. We provide a taxonomy based on the core properties of the watermark. We discuss the functionality and security threats of AIGC watermarking.
  arXiv  Detail & Related papers  (2024-11-18T11:26:42Z)
• Beyond Dataset Watermarking: Model-Level Copyright Protection for Code Summarization Models [37.817691840557984]
  CSMs face risks of exploitation by unauthorized users. Traditional watermarking methods require separate design of triggers and watermark features. We propose ModMark, a novel model-level digital watermark embedding method.
  arXiv  Detail & Related papers  (2024-10-18T00:48:00Z)
• Certifiably Robust Image Watermark [57.546016845801134]
  Generative AI raises many societal concerns such as boosting disinformation and propaganda campaigns. Watermarking AI-generated content is a key technology to address these concerns. We propose the first image watermarks with certified robustness guarantees against removal and forgery attacks.
  arXiv  Detail & Related papers  (2024-07-04T17:56:04Z)
• Uncovering LLM-Generated Code: A Zero-Shot Synthetic Code Detector via Code Rewriting [78.48355455324688]
  We propose a novel zero-shot synthetic code detector based on the similarity between the original code and its LLM-rewritten variants.<n>Our results demonstrate a significant improvement over existing SOTA synthetic content detectors.
  arXiv  Detail & Related papers  (2024-05-25T08:57:28Z)
• CodeIP: A Grammar-Guided Multi-Bit Watermark for Large Language Models of Code [56.019447113206006]
  Large Language Models (LLMs) have achieved remarkable progress in code generation. CodeIP is a novel multi-bit watermarking technique that embeds additional information to preserve provenance details. Experiments conducted on a real-world dataset across five programming languages demonstrate the effectiveness of CodeIP.
  arXiv  Detail & Related papers  (2024-04-24T04:25:04Z)
• Is The Watermarking Of LLM-Generated Code Robust? [5.48277165801539]
  We show that watermarking techniques are significantly more fragile in code-based contexts. Specifically, we show that simple semantic-preserving transformations, such as variable renaming and dead code insertion, can effectively erase watermarks.
  arXiv  Detail & Related papers  (2024-03-24T21:41:29Z)
• No Free Lunch in LLM Watermarking: Trade-offs in Watermarking Design Choices [20.20770405297239]
  We show that common design choices in LLM watermarking schemes make the resulting systems surprisingly susceptible to attack. We propose guidelines and defenses for LLM watermarking in practice.
  arXiv  Detail & Related papers  (2024-02-25T20:24:07Z)
• WatME: Towards Lossless Watermarking Through Lexical Redundancy [58.61972059246715]
  This study assesses the impact of watermarking on different capabilities of large language models (LLMs) from a cognitive science lens. We introduce Watermarking with Mutual Exclusion (WatME) to seamlessly integrate watermarks.
  arXiv  Detail & Related papers  (2023-11-16T11:58:31Z)
• A Robust Semantics-based Watermark for Large Language Model against Paraphrasing [50.84892876636013]
  Large language models (LLMs) have show great ability in various natural language tasks. There are concerns that LLMs are possible to be used improperly or even illegally. We propose a semantics-based watermark framework SemaMark.
  arXiv  Detail & Related papers  (2023-11-15T06:19:02Z)
• An Unforgeable Publicly Verifiable Watermark for Large Language Models [84.2805275589553]
  Current watermark detection algorithms require the secret key used in the watermark generation process, making them susceptible to security breaches and counterfeiting during public detection. We propose an unforgeable publicly verifiable watermark algorithm named UPV that uses two different neural networks for watermark generation and detection, instead of using the same key at both stages.
  arXiv  Detail & Related papers  (2023-07-30T13:43:27Z)
• Who Wrote this Code? Watermarking for Code Generation [53.24895162874416]
  We propose Selective WatErmarking via Entropy Thresholding (SWEET) to detect machine-generated text. Our experiments show that SWEET significantly improves code quality preservation while outperforming all baselines.
  arXiv  Detail & Related papers  (2023-05-24T11:49:52Z)
• Evading Watermark based Detection of AI-Generated Content [45.47476727209842]
  A generative AI model can generate extremely realistic-looking content. Watermark has been leveraged to detect AI-generated content. A content is detected as AI-generated if a similar watermark can be decoded from it.
  arXiv  Detail & Related papers  (2023-05-05T19:20:29Z)
• Generation Probabilities Are Not Enough: Uncertainty Highlighting in AI Code Completions [54.55334589363247]
  We study whether conveying information about uncertainty enables programmers to more quickly and accurately produce code. We find that highlighting tokens with the highest predicted likelihood of being edited leads to faster task completion and more targeted edits.
  arXiv  Detail & Related papers  (2023-02-14T18:43:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.