# Protecting Creative Writing Copyright against AI Imitation via Implicit Watermarking

arXiv: https://arxiv.org/abs/2504.00035 · HTML: https://mlsanity.com/paper/2504.00035/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Watermarking generated content and models, one of 12 directions in LLM safety and security.
- Builds on 12 of its 37 references that are in the map.
- First posted on arXiv 30 Mar 2025.

## 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, 12 of 12)

- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [OPT: Open Pre-trained Transformer Language Models](https://mlsanity.com/paper/2205.01068/) (2022)
- [SimCSE: Simple Contrastive Learning of Sentence Embeddings](https://mlsanity.com/paper/2104.08821/) (2021)
- [Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey](https://mlsanity.com/paper/2403.14608/) (2024)
- [Robust Distortion-free Watermarks for Language Models](https://mlsanity.com/paper/2307.15593/) (2023)
- [Raising the Cost of Malicious AI-Powered Image Editing](https://mlsanity.com/paper/2302.06588/) (2023)
- [Style Transformer: Unpaired Text Style Transfer without Disentangled Latent Representation](https://mlsanity.com/paper/1905.05621/) (2019)
- [Protecting Intellectual Property of Language Generation APIs with Lexical Watermark](https://mlsanity.com/paper/2112.02701/) (2021)
- [An Entropy-based Text Watermarking Detection Method](https://mlsanity.com/paper/2403.13485/) (2024)
- [Watermarking Text Data on Large Language Models for Dataset Copyright](https://mlsanity.com/paper/2305.13257/) (2023)
- [StoryTrans: Non-Parallel Story Author-Style Transfer with Discourse Representations and Content Enhancing](https://mlsanity.com/paper/2208.13423/) (2022)

## Abstract (arXiv)

Large language models (LLMs) enable powerful knowledge injection through approaches such as in-context learning and fine-tuning, but they also introduce new risks of unauthorized imitation of high-value creative works. Existing copyright protection techniques mainly focus on visual media, leaving the protection of creative writing largely unexplored. In this work, we investigate a new challenge: verifying whether AI-generated texts inherit the creative essence of protected works without authorization. We propose WIND (Watermarking via Implicit and Non-disruptive Disentanglement), a zero-watermarking framework that constructs an implicit and verifiable creative signature for copyright verification. WIND decomposes creative essence into five complementary dimensions and leverages an LLM-based instance delimitation mechanism to extract condensed representations of protected creative characteristics. By disentangling creative-specific information from irrelevant textual variations, these representations are mapped into a compact watermark space without modifying the original texts. Extensive experiments demonstrate that WIND achieves over 98% F1 scores while maintaining low false-positive rates, substantially outperforming existing watermarking and text classification approaches under various AI imitation scenarios.