---
title: "Rethinking LLM Watermark Detection in Black-Box Settings: A"
description: "The authors propose TTP-Detect, a black-box framework that lets an accredited third-party auditor… 4% AUC on SymMark. 40 references."
authors: ["Zhuoshang Wang", "Yubing Ren", "Yanan Cao", "Fang Fang", "Xiaoxue Li", "Li Guo"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: ACL 2026"
source: "https://aclanthology.org/2026.findings-acl.990/"
doi: "10.18653/v1/2026.findings-acl.990"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/wang-2026-rethinking-llm-watermark-detection-black/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/wang-2026-rethinking-llm-watermark-detection-black/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/47/) / Rethinking LLM Watermark Detection in Black-Box Settings: A Non-Intrusive Third-Party Framework

# Rethinking LLM Watermark Detection in Black-Box Settings: A Non-Intrusive Third-Party Framework

**Authors:** Zhuoshang Wang, [Yubing Ren](https://md.research.elamur.ai/free/en/authors/yubing-ren/), [Yanan Cao](https://md.research.elamur.ai/free/en/authors/yanan-cao/), [Fang Fang](https://md.research.elamur.ai/free/en/authors/fang-fang/), Xiaoxue Li and [Li Guo](https://md.research.elamur.ai/free/en/authors/li-guo/) · **Year:** 2026 · **Venue:** [Findings of the Association for Computational Linguistics: ACL 2026](https://md.research.elamur.ai/free/en/venues/findings/) ([Findings of the Association for Computational Linguistics 2026](https://md.research.elamur.ai/free/en/venues/findings/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.findings-acl.990/) · [PDF](https://aclanthology.org/2026.findings-acl.990.pdf) · [DOI 10.18653/v1/2026.findings-acl.990](https://doi.org/10.18653/v1/2026.findings-acl.990) · [OpenAlex](https://openalex.org/W7166881387)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
Most LLM watermarks share a secret key between the embedding step and the detector. Outsiders such as courts, regulators or platform moderators therefore cannot check a watermark on their own. They must either trust the provider's word or obtain the key, and handing over the key makes forging or removing the watermark easier.

The authors propose TTP-Detect, a black-box framework that lets an accredited third-party auditor verify watermarks without keys and without knowing the scheme. The auditor uses the provider's ordinary generation API, switching a watermark flag on and off, to collect small reference sets of watermarked and unwatermarked outputs. Detection then becomes a relative hypothesis test: is the query text closer to the watermarked set than to the clean one? Texts are embedded with a fine-tuned proxy model that sharpens watermark cues. Several complementary relative measurements are then combined into one calibrated score.

Across many logits- and sampling-based schemes, the authors report strong detection and robustness to attacks, with low auditor-side cost. This points toward neutral watermark auditing that does not depend on any one scheme.

## Key points
- **Four relative measurement modules** are ensembled: a Local Consistency Test, a Relative Mahalanobis Score, a Relative Energy Score, and an Adaptive Rank Test based on post-hoc likelihood re-scoring. In leave-one-out ablation on SynthID, removing the Adaptive Rank Test caused the largest AUC losses.
- **Evaluated schemes and models:** KGW, SWEET, Unigram, SymMark, SynthID, MorphMark and Unbiased. The provider-side generator was Llama-3.1-8B, and Qwen2.5-3B served as the proxy for representations.
- **Proxy fine-tuning is helpful but not required.** An untuned Qwen2.5-3B-Base proxy still worked, losing under 4% AUC on SymMark and Unigram.
- **Reference set size:** N=16 samples per class was chosen as the default. For distribution-preserving schemes (Unbiased, SynthID), using N=8 and N=16 lifted the mean best F1 relative to the reference-free setting, with gains exceeding 5% and 10% respectively. Gains flattened beyond N=16.
- **Cost:** generating the reference samples on the provider side took about 6.14 s. All auditor-side computation finished in under 2 s. The authors note that schemes designed to minimize distortion leave a weaker detectable signal.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Chinese Academy of Sciences](https://md.research.elamur.ai/free/en/government/chinese-academy-of-sciences/) — Government body, Beijing, China · 246 works · 24,952 citations
- [University of Chinese Academy of Sciences](https://md.research.elamur.ai/free/en/universities/university-of-chinese-academy-of-sciences/) — University, Beijing, China · 188 works · 8,744 citations
- [Institute of Information Engineering](https://md.research.elamur.ai/free/en/labs/institute-of-information-engineering/) — Research institute, Beijing, China · 41 works · 1,485 citations
- [National Computer Network Emergency Response Technical Team/Coordination Center of Chinar](https://md.research.elamur.ai/free/en/ngos/national-computer-network-emergency-response-technical-team-coordination-center/) — Nonprofit, Beijing, China · 2 works

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 3,486 works · [🔌 Hardware and Architecture](https://md.research.elamur.ai/free/en/shelves/computer-science/hardware-and-architecture/) · 167 works

## Cite

Cite as: Zhuoshang Wang, Yubing Ren, Yanan Cao, Fang Fang, Xiaoxue Li and Li Guo. 2026. Rethinking LLM Watermark Detection in Black-Box Settings: A Non-Intrusive Third-Party Framework. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{wang2026rethinking,
  title = {{Rethinking LLM Watermark Detection in Black-Box Settings: A Non-Intrusive Third-Party Framework}},
  author = {Zhuoshang Wang, Yubing Ren, Yanan Cao, Fang Fang, Xiaoxue Li and Li Guo},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.990},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.990}
}
```

- [← previous Rethinking Gender Annotation for Bias Evaluation in Machine Translation: Can LLMs Improve Reliability?](https://md.research.elamur.ai/free/en/papers/manna-2026-rethinking-gender-annotation-bias-evaluation/)
- [next → Rethinking Reading Order: Toward Generalizable Document Understanding with LLM-based Relation Modeling](https://md.research.elamur.ai/free/en/papers/wang-2026-rethinking-reading-order-generalizable-document/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/47/)

## References

40 references, 15 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Scott Aaronson. 2023. Watermarking of large language models. Talk at the Large Language Models and Transformers Workshop, Simons Institute for the Theory of Computing. [find](https://scholar.google.com/scholar?q=Scott+Aaronson%2E+2023%2E+Watermarking+of+large+language+models%2E+Talk+at+the+Large+Language+Models+and+Transformers+Workshop%2C+Simons+Institute+for+the+Theory+of+Computing%2E)
2. Marah Abdin, Jyoti Aneja, Hany Awadalla, Ahmed Awadallah, Ammar Ahmad Awan, Nguyen Bach, Amit Bahree, Arash Bakhtiari, Jianmin Bao, Harkirat Behl, Alon Benhaim, Misha Bilenko, Johan Bjorck, Sébastien Bubeck, Martin Cai, Qin Cai, Vishrav Chaudhary, Dong Chen, Dongdong Chen, and 110 others. 2024. Phi-… — [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://md.research.elamur.ai/free/en/papers/abdin-2024-phi-3-technical-report-highly/)
3. Ruibo Chen, Yihan Wu, Junfeng Guo, and Heng Huang. 2025. Improved unbiased watermark for large language models. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 20587– 20601, Vienna, Austria. Association for Computational Lingu… [find](https://scholar.google.com/scholar?q=Ruibo+Chen%2C+Yihan+Wu%2C+Junfeng+Guo%2C+and+Heng+Huang%2E+2025%2E+Improved+unbiased+watermark+for+large+language+models%2E+In+Proceedings+of+the+63rd+Annual+Meeting+of+the+Association+for+Computational+Linguisti)
4. Miranda Christ, Sam Gunn, and Or Zamir. 2024. Undetectable watermarks for language models. In Proceedings of Thirty Seventh Conference on Learning Theory, volume 247 of Proceedings of Machine Learning Research, pages 1125–1139. PMLR. [find](https://scholar.google.com/scholar?q=Miranda+Christ%2C+Sam+Gunn%2C+and+Or+Zamir%2E+2024%2E+Undetectable+watermarks+for+language+models%2E+In+Proceedings+of+Thirty+Seventh+Conference+on+Learning+Theory%2C+volume+247+of+Proceedings+of+Machine+Learning)
5. Amirhossein Dabiriaghdam and Lele Wang. 2025. Sim-Mark: A robust sentence-level similarity-based watermarking algorithm for large language models. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 30773–30794, Suzhou, China. Association for Computationa… [find](https://scholar.google.com/scholar?q=Amirhossein+Dabiriaghdam+and+Lele+Wang%2E+2025%2E+Sim%2DMark%3A+A+robust+sentence%2Dlevel+similarity%2Dbased+watermarking+algorithm+for+large+language+models%2E+In+Proceedings+of+the+2025+Conference+on+Empirical+Me)
6. Sumanth Dathathri, Abigail See, Sumedh Ghaisas, Po-Sen Huang, Rob McAdam, Johannes Welbl, Vandana Bachani, Alex Kaskasoli, Robert Stanforth, Tatiana Matejovicova, and 1 others. 2024. Scalable watermarking for identifying large language model outputs. Nature, 634(8035):818–823. [find](https://scholar.google.com/scholar?q=Sumanth+Dathathri%2C+Abigail+See%2C+Sumedh+Ghaisas%2C+Po%2DSen+Huang%2C+Rob+McAdam%2C+Johannes+Welbl%2C+Vandana+Bachani%2C+Alex+Kaskasoli%2C+Robert+Stanforth%2C+Tatiana+Matejovicova%2C+and+1+others%2E+2024%2E+Scalable+watermar)
7. Haohua Duan, Liyao Xiang, and Xin Zhang. 2025. Pvmark: Enabling public verifiability for llm watermarking schemes. Preprint, arXiv:2510.26274. — [arXiv:2510.26274](https://arxiv.org/abs/2510.26274) [find](https://scholar.google.com/scholar?q=Haohua+Duan%2C+Liyao+Xiang%2C+and+Xin+Zhang%2E+2025%2E+Pvmark%3A+Enabling+public+verifiability+for+llm+watermarking+schemes%2E+Preprint%2C+arXiv%3A2510%2E26274%2E+%E2%80%94+%5BarXiv%3A2510%2E26274%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2510%2E26274%29)
8. Jaiden Fairoze, Sanjam Garg, Somesh Jha, Saeed Mahloujifar, Mohammad Mahmoody, and Mingyuan Wang. 2025. Publicly-detectable watermarking for language models. Preprint, arXiv:2310.18491. — [arXiv:2310.18491](https://arxiv.org/abs/2310.18491) [find](https://scholar.google.com/scholar?q=Jaiden+Fairoze%2C+Sanjam+Garg%2C+Somesh+Jha%2C+Saeed+Mahloujifar%2C+Mohammad+Mahmoody%2C+and+Mingyuan+Wang%2E+2025%2E+Publicly%2Ddetectable+watermarking+for+language+models%2E+Preprint%2C+arXiv%3A2310%2E18491%2E+%E2%80%94+%5BarXiv%3A2310%2E)
9. Angela Fan, Yacine Jernite, Ethan Perez, David Grangier, Jason Weston, and Michael Auli. 2019. [ELI5: Long form question answering](https://md.research.elamur.ai/free/en/papers/fan-2019-eli5-long-form-question-answering/). In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 3558–3567, Florence, Italy. Association for Computational Linguistics.
10. Jiayi Fu, Xuandong Zhao, Ruihan Yang, Yuansen Zhang, Jiangjie Chen, and Yanghua Xiao. 2024. Gumbel-Soft: Diversified language model watermarking via the GumbelMax-trick. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 5791– 58… [find](https://scholar.google.com/scholar?q=Jiayi+Fu%2C+Xuandong+Zhao%2C+Ruihan+Yang%2C+Yuansen+Zhang%2C+Jiangjie+Chen%2C+and+Yanghua+Xiao%2E+2024%2E+Gumbel%2DSoft%3A+Diversified+language+model+watermarking+via+the+GumbelMax%2Dtrick%2E+In+Proceedings+of+the+62nd+Ann)
11. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, Amy Yang, Angela Fan, Anirudh Goyal, Anthony Hartshorn, Aobo Yang, Archi Mitra, Archie Sravankumar, Artem Korenev, Arthur Hinsvark, and 542 o… — [The Llama 3 Herd of Models](https://md.research.elamur.ai/free/en/papers/grattafiori-2024-llama-3-herd-models/)
12. Chenchen Gu, Xiang Lisa Li, Percy Liang, and Tatsunori Hashimoto. 2024. On the learnability of watermarks for language models. In The Twelfth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Chenchen+Gu%2C+Xiang+Lisa+Li%2C+Percy+Liang%2C+and+Tatsunori+Hashimoto%2E+2024%2E+On+the+learnability+of+watermarks+for+language+models%2E+In+The+Twelfth+International+Conference+on+Learning+Representations%2E)
13. Zhiwei He, Binglin Zhou, Hongkun Hao, Aiwei Liu, Xing Wang, Zhaopeng Tu, Zhuosheng Zhang, and Rui Wang. 2024. Can watermarks survive translation? on the cross-lingual consistency of text watermark for large language models. In Proceedings of the 62nd Annual Meeting of the Association for Computation… [find](https://scholar.google.com/scholar?q=Zhiwei+He%2C+Binglin+Zhou%2C+Hongkun+Hao%2C+Aiwei+Liu%2C+Xing+Wang%2C+Zhaopeng+Tu%2C+Zhuosheng+Zhang%2C+and+Rui+Wang%2E+2024%2E+Can+watermarks+survive+translation%3F+on+the+cross%2Dlingual+consistency+of+text+watermark+for)
14. Abe Hou, Jingyu Zhang, Tianxing He, Yichen Wang, Yung-Sung Chuang, Hongwei Wang, Lingfeng Shen, Benjamin Van Durme, Daniel Khashabi, and Yulia Tsvetkov. 2024a. SemStamp: A semantic watermark with paraphrastic robustness for text generation. In Proceedings of the 2024 Conference of the North American… [find](https://scholar.google.com/scholar?q=Abe+Hou%2C+Jingyu+Zhang%2C+Tianxing+He%2C+Yichen+Wang%2C+Yung%2DSung+Chuang%2C+Hongwei+Wang%2C+Lingfeng+Shen%2C+Benjamin+Van+Durme%2C+Daniel+Khashabi%2C+and+Yulia+Tsvetkov%2E+2024a%2E+SemStamp%3A+A+semantic+watermark+with+para)
15. Abe Hou, Jingyu Zhang, Yichen Wang, Daniel Khashabi, and Tianxing He. 2024b. k-SemStamp: A clustering-based semantic watermark for detection of machine-generated text. In Findings of the Association for Computational Linguistics: ACL 2024, pages 1706–1715, Bangkok, Thailand. Association for Computat… [find](https://scholar.google.com/scholar?q=Abe+Hou%2C+Jingyu+Zhang%2C+Yichen+Wang%2C+Daniel+Khashabi%2C+and+Tianxing+He%2E+2024b%2E+k%2DSemStamp%3A+A+clustering%2Dbased+semantic+watermark+for+detection+of+machine%2Dgenerated+text%2E+In+Findings+of+the+Association+f)
16. Edward J Hu, yelong shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2022. [LoRA: Low-rank adaptation of large language models](https://md.research.elamur.ai/free/en/papers/hu-2021-lora-low-rank-adaptation-large/). In International Conference on Learning Representations.
17. Zhengmian Hu, Lichang Chen, Xidong Wu, Yihan Wu, Hongyang Zhang, and Heng Huang. 2024. Unbiased watermark for large language models. In The Twelfth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Zhengmian+Hu%2C+Lichang+Chen%2C+Xidong+Wu%2C+Yihan+Wu%2C+Hongyang+Zhang%2C+and+Heng+Huang%2E+2024%2E+Unbiased+watermark+for+large+language+models%2E+In+The+Twelfth+International+Conference+on+Learning+Representations)
18. Mingjia Huo, Sai Ashish Somayajula, Youwei Liang, Ruisi Zhang, Farinaz Koushanfar, and Pengtao Xie. 2024. Token-specific watermarking with enhanced detectability and semantic coherence for large language models. In Proceedings of the 41st International Conference on Machine Learning, ICML’24. JMLR.o… [find](https://scholar.google.com/scholar?q=Mingjia+Huo%2C+Sai+Ashish+Somayajula%2C+Youwei+Liang%2C+Ruisi+Zhang%2C+Farinaz+Koushanfar%2C+and+Pengtao+Xie%2E+2024%2E+Token%2Dspecific+watermarking+with+enhanced+detectability+and+semantic+coherence+for+large+langu)
19. John Kirchenbauer, Jonas Geiping, Yuxin Wen, Jonathan Katz, Ian Miers, and Tom Goldstein. 2023. [A watermark for large language models](https://md.research.elamur.ai/free/en/papers/kirchenbauer-2023-watermark-large-language-models/). In Proceedings of the 40th International Conference on Machine Learning, volume 202 of Proceedings of Machine Learning Research, pages 17061–17084. PMLR.
20. Kalpesh Krishna, Yixiao Song, Marzena Karpinska, John Frederick Wieting, and Mohit Iyyer. 2023. [Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defense](https://md.research.elamur.ai/free/en/papers/krishna-2023-paraphrasing-evades-detectors-ai-generated/). In Thirty-seventh Conference on Neural Information Processing Systems.
21. Rohith Kuditipudi, John Thickstun, Tatsunori Hashimoto, and Percy Liang. 2024. [Robust distortion-free watermarks for language models](https://md.research.elamur.ai/free/en/papers/kuditipudi-2023-robust-distortion-free-watermarks-language/). Transactions on Machine Learning Research.
22. Taehyun Lee, Seokhee Hong, Jaewoo Ahn, Ilgee Hong, Hwaran Lee, Sangdoo Yun, Jamin Shin, and Gunhee Kim. 2024. [Who wrote this code? watermarking for code generation](https://md.research.elamur.ai/free/en/papers/lee-2024-who-wrote-this-code-watermarking/). In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 4890–4911, Ba…
23. Aiwei Liu, Leyi Pan, Xuming Hu, Shuang Li, Lijie Wen, Irwin King, and Philip S. Yu. 2024a. An unforgeable publicly verifiable watermark for large language models. In The Twelfth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Aiwei+Liu%2C+Leyi+Pan%2C+Xuming+Hu%2C+Shuang+Li%2C+Lijie+Wen%2C+Irwin+King%2C+and+Philip+S%2E+Yu%2E+2024a%2E+An+unforgeable+publicly+verifiable+watermark+for+large+language+models%2E+In+The+Twelfth+International+Conferen)
24. Aiwei Liu, Leyi Pan, Xuming Hu, Shiao Meng, and Lijie Wen. 2024b. A semantic invariant robust watermark for large language models. In The Twelfth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Aiwei+Liu%2C+Leyi+Pan%2C+Xuming+Hu%2C+Shiao+Meng%2C+and+Lijie+Wen%2E+2024b%2E+A+semantic+invariant+robust+watermark+for+large+language+models%2E+In+The+Twelfth+International+Conference+on+Learning+Representations%2E)
25. Yepeng Liu and Yuheng Bu. 2024. Adaptive text watermark for large language models. In Proceedings of the 41st International Conference on Machine Learning, ICML’24. JMLR.org. [find](https://scholar.google.com/scholar?q=Yepeng+Liu+and+Yuheng+Bu%2E+2024%2E+Adaptive+text+watermark+for+large+language+models%2E+In+Proceedings+of+the+41st+International+Conference+on+Machine+Learning%2C+ICML%E2%80%9924%2E+JMLR%2Eorg%2E)
26. Yijian Lu, Aiwei Liu, Dianzhi Yu, Jingjing Li, and Irwin King. 2024. An entropy-based text watermarking detection method. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 11724– 11735, Bangkok, Thailand. Association for Computa… [find](https://scholar.google.com/scholar?q=Yijian+Lu%2C+Aiwei+Liu%2C+Dianzhi+Yu%2C+Jingjing+Li%2C+and+Irwin+King%2E+2024%2E+An+entropy%2Dbased+text+watermarking+detection+method%2E+In+Proceedings+of+the+62nd+Annual+Meeting+of+the+Association+for+Computational)
27. Minjia Mao, Dongjun Wei, Zeyu Chen, Xiao Fang, and Michael Chau. 2025. Watermarking large language models: An unbiased and low-risk method. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 7939–7960, Vienna, Austria. Associatio… [find](https://scholar.google.com/scholar?q=Minjia+Mao%2C+Dongjun+Wei%2C+Zeyu+Chen%2C+Xiao+Fang%2C+and+Michael+Chau%2E+2025%2E+Watermarking+large+language+models%3A+An+unbiased+and+low%2Drisk+method%2E+In+Proceedings+of+the+63rd+Annual+Meeting+of+the+Association)
28. George A. Miller. 1995. [Wordnet: a lexical database for english](https://md.research.elamur.ai/free/en/papers/miller-1994-wordnet/). Commun. ACM, 38(11):39–41.
29. OpenAI. 2025. GPT-5.1 Model (OpenAI API Documentation). [https://platform.openai.com/docs/](https://platform.openai.com/docs/) models/gpt-5.1. [↗](https://platform.openai.com/docs/)
30. Leyi Pan, Aiwei Liu, Zhiwei He, Zitian Gao, Xuandong Zhao, Yijian Lu, Binglin Zhou, Shuliang Liu, Xuming Hu, Lijie Wen, Irwin King, and Philip S. Yu. 2024. MarkLLM: An open-source toolkit for LLM watermarking. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing:… [find](https://scholar.google.com/scholar?q=Leyi+Pan%2C+Aiwei+Liu%2C+Zhiwei+He%2C+Zitian+Gao%2C+Xuandong+Zhao%2C+Yijian+Lu%2C+Binglin+Zhou%2C+Shuliang+Liu%2C+Xuming+Hu%2C+Lijie+Wen%2C+Irwin+King%2C+and+Philip+S%2E+Yu%2E+2024%2E+MarkLLM%3A+An+open%2Dsource+toolkit+for+LLM+wate)
31. Qwen, An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, Huan Lin, Jian Yang, Jianhong Tu, Jianwei Zhang, Jianxin Yang, Jiaxi Yang, Jingren Zhou, Junyang Lin, and 24 others. 2025. Qwen2.5 technical report. Preprint, arXiv:2412.15… [find](https://scholar.google.com/scholar?q=Qwen%2C+An+Yang%2C+Baosong+Yang%2C+Beichen+Zhang%2C+Binyuan+Hui%2C+Bo+Zheng%2C+Bowen+Yu%2C+Chengyuan+Li%2C+Dayiheng+Liu%2C+Fei+Huang%2C+Haoran+Wei%2C+Huan+Lin%2C+Jian+Yang%2C+Jianhong+Tu%2C+Jianwei+Zhang%2C+Jianxin+Yang%2C+Jiaxi+Yan)
32. Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J. Liu. 2020. [Exploring the limits of transfer learning with a unified text-to-text transformer](https://md.research.elamur.ai/free/en/papers/raffel-2019-exploring-limits-transfer-learning-unified/). J. Mach. Learn. Res., 21(1).
33. Jie Ren, Han Xu, Yiding Liu, Yingqian Cui, Shuaiqiang Wang, Dawei Yin, and Jiliang Tang. 2024a. A robust semantics-based watermark for large language model against paraphrasing. In Findings of the Association for Computational Linguistics: NAACL 2024, pages 613–625, Mexico City, Mexico. Association… [find](https://scholar.google.com/scholar?q=Jie+Ren%2C+Han+Xu%2C+Yiding+Liu%2C+Yingqian+Cui%2C+Shuaiqiang+Wang%2C+Dawei+Yin%2C+and+Jiliang+Tang%2E+2024a%2E+A+robust+semantics%2Dbased+watermark+for+large+language+model+against+paraphrasing%2E+In+Findings+of+the+Ass)
34. Yubing Ren, Ping Guo, Yanan Cao, and Wei Ma. 2024b. Subtle signatures, strong shields: Advancing robust and imperceptible watermarking in large language models. In Findings of the Association for Computational Linguistics: ACL 2024, pages 5508–5519, Bangkok, Thailand. Association for Computational L… [find](https://scholar.google.com/scholar?q=Yubing+Ren%2C+Ping+Guo%2C+Yanan+Cao%2C+and+Wei+Ma%2E+2024b%2E+Subtle+signatures%2C+strong+shields%3A+Advancing+robust+and+imperceptible+watermarking+in+large+language+models%2E+In+Findings+of+the+Association+for+Comp)
35. Yidan Wang, Yubing Ren, Yanan Cao, and Binxing Fang. 2025a. From trade-off to synergy: A versatile symbiotic watermarking framework for large language models. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 10306– 10322, Vienn… [find](https://scholar.google.com/scholar?q=Yidan+Wang%2C+Yubing+Ren%2C+Yanan+Cao%2C+and+Binxing+Fang%2E+2025a%2E+From+trade%2Doff+to+synergy%3A+A+versatile+symbiotic+watermarking+framework+for+large+language+models%2E+In+Proceedings+of+the+63rd+Annual+Meeting)
36. Zongqi Wang, Tianle Gu, Baoyuan Wu, and Yujiu Yang. 2025b. [MorphMark: Flexible adaptive watermarking for large language models](https://md.research.elamur.ai/free/en/papers/wang-2025-morphmark-flexible-adaptive-watermarking-large/). In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 4842–4860, Vienna, Austria. Association for Comput…
37. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, Chujie Zheng, Dayiheng Liu, Fan Zhou, Fei Huang, Feng Hu, Hao Ge, Haoran Wei, Huan Lin, Jialong Tang, and 41 others. 2025. [Qwen3 technical report](https://md.research.elamur.ai/free/en/papers/yang-2025-qwen3-technical-report/). Preprint, arXiv:2505.09388.
38. Susan Zhang, Stephen Roller, Naman Goyal, Mikel Artetxe, Moya Chen, Shuohui Chen, Christopher Dewan, Mona Diab, Xian Li, Xi Victoria Lin, Todor Mihaylov, Myle Ott, Sam Shleifer, Kurt Shuster, Daniel Simig, Punit Singh Koura, Anjali Sridhar, Tianlu Wang, and Luke Zettlemoyer. 2022. [Opt: Open pre-trai…](https://md.research.elamur.ai/free/en/papers/zhang-2022-opt-open-pre-trained-transformer/)
39. Xuandong Zhao, Prabhanjan Vijendra Ananth, Lei Li, and Yu-Xiang Wang. 2024. [Provable robust watermarking for AI-generated text](https://md.research.elamur.ai/free/en/papers/zhao-2023-provable-robust-watermarking-ai-generated/). In The Twelfth International Conference on Learning Representations.
40. Yaowei Zheng, Richong Zhang, Junhao Zhang, Yanhan Ye, Zheyan Luo, Zhangchi Feng, and Yongqiang Ma. 2024. [Llamafactory: Unified efficient fine-tuning of 100+ language models](https://md.research.elamur.ai/free/en/papers/zheng-2024-llamafactory-unified-efficient-fine-tuning/). In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 3: System Demonstrations),…

*Updated 2 November 2026*