{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,25]],"date-time":"2026-04-25T15:22:19Z","timestamp":1777130539474,"version":"3.51.4"},"reference-count":47,"publisher":"MDPI AG","issue":"8","license":[{"start":{"date-parts":[[2025,7,24]],"date-time":"2025-07-24T00:00:00Z","timestamp":1753315200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"National Natural Science Foundation of China","award":["42301484"],"award-info":[{"award-number":["42301484"]}]},{"name":"National Natural Science Foundation of China","award":["HNGISA2023004"],"award-info":[{"award-number":["HNGISA2023004"]}]},{"name":"National Natural Science Foundation of China","award":["JGY2023069"],"award-info":[{"award-number":["JGY2023069"]}]},{"name":"National Natural Science Foundation of China","award":["202410327136Y"],"award-info":[{"award-number":["202410327136Y"]}]},{"name":"Open Topic of Hunan Engineering Research Center of Geographic Information Security and Application","award":["42301484"],"award-info":[{"award-number":["42301484"]}]},{"name":"Open Topic of Hunan Engineering Research Center of Geographic Information Security and Application","award":["HNGISA2023004"],"award-info":[{"award-number":["HNGISA2023004"]}]},{"name":"Open Topic of Hunan Engineering Research Center of Geographic Information Security and Application","award":["JGY2023069"],"award-info":[{"award-number":["JGY2023069"]}]},{"name":"Open Topic of Hunan Engineering Research Center of Geographic Information Security and Application","award":["202410327136Y"],"award-info":[{"award-number":["202410327136Y"]}]},{"name":"Teaching Reform Project of Nanjing University of Finance and Economics","award":["42301484"],"award-info":[{"award-number":["42301484"]}]},{"name":"Teaching Reform Project of Nanjing University of Finance and Economics","award":["HNGISA2023004"],"award-info":[{"award-number":["HNGISA2023004"]}]},{"name":"Teaching Reform Project of Nanjing University of Finance and Economics","award":["JGY2023069"],"award-info":[{"award-number":["JGY2023069"]}]},{"name":"Teaching Reform Project of Nanjing University of Finance and Economics","award":["202410327136Y"],"award-info":[{"award-number":["202410327136Y"]}]},{"name":"Provincial Undergraduate Training Program on Innovation and Entrepreneurship","award":["42301484"],"award-info":[{"award-number":["42301484"]}]},{"name":"Provincial Undergraduate Training Program on Innovation and Entrepreneurship","award":["HNGISA2023004"],"award-info":[{"award-number":["HNGISA2023004"]}]},{"name":"Provincial Undergraduate Training Program on Innovation and Entrepreneurship","award":["JGY2023069"],"award-info":[{"award-number":["JGY2023069"]}]},{"name":"Provincial Undergraduate Training Program on Innovation and Entrepreneurship","award":["202410327136Y"],"award-info":[{"award-number":["202410327136Y"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Entropy"],"abstract":"<jats:p>Large language models (LLMs) pose significant challenges to content authentication, as their sophisticated generation capabilities make distinguishing AI-produced text from human writing increasingly difficult. Current detection methods suffer from limited information capture, poor rate\u2013distortion trade-offs, and vulnerability to adversarial perturbations. We present CurveMark, a novel dual-channel detection framework that combines probability curvature analysis with dynamic semantic watermarking, grounded in information-theoretic principles to maximize mutual information between text sources and observable features. To address the limitation of requiring prior knowledge of source models, we incorporate a Bayesian multi-hypothesis detection framework for statistical inference without prior assumptions. Our approach embeds imperceptible watermarks during generation via entropy-aware, semantically informed token selection and extracts complementary features from probability curvature patterns and watermark-specific metrics. Evaluation across multiple datasets and LLM architectures demonstrates 95.4% detection accuracy with minimal quality degradation (perplexity increase &lt; 1.3), achieving 85\u201389% channel capacity utilization and robust performance under adversarial perturbations (72\u201394% information retention).<\/jats:p>","DOI":"10.3390\/e27080784","type":"journal-article","created":{"date-parts":[[2025,7,24]],"date-time":"2025-07-24T14:11:44Z","timestamp":1753366304000},"page":"784","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":4,"title":["CurveMark: Detecting AI-Generated Text via Probabilistic Curvature and Dynamic Semantic Watermarking"],"prefix":"10.3390","volume":"27","author":[{"given":"Yuhan","family":"Zhang","sequence":"first","affiliation":[{"name":"School of Computer Science and Artificial Intelligence, Nanjing University of Finance and Economics, Nanjing 210023, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xingxiang","family":"Jiang","sequence":"additional","affiliation":[{"name":"The Third Surveying and Mapping Institute of Hunan Province, Changsha 410018, China"},{"name":"Hunan Engineering Research Center of Geographic Information Security and Application, Changsha 410007, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hua","family":"Sun","sequence":"additional","affiliation":[{"name":"The Third Surveying and Mapping Institute of Hunan Province, Changsha 410018, China"},{"name":"Hunan Engineering Research Center of Geographic Information Security and Application, Changsha 410007, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yao","family":"Zhang","sequence":"additional","affiliation":[{"name":"The Third Surveying and Mapping Institute of Hunan Province, Changsha 410018, China"},{"name":"Hunan Engineering Research Center of Geographic Information Security and Application, Changsha 410007, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1009-750X","authenticated-orcid":false,"given":"Deyu","family":"Tong","sequence":"additional","affiliation":[{"name":"School of Computer Science and Artificial Intelligence, Nanjing University of Finance and Economics, Nanjing 210023, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2025,7,24]]},"reference":[{"key":"ref_1","unstructured":"Bakhtin, A., Gross, S., Ott, M., Deng, Y., Ranzato, M., and Szlam, A. (2019). Real or Fake? Learning to Discriminate Machine from Human Generated Text. arXiv."},{"key":"ref_2","first-page":"1","article-title":"Defending Against Neural Fake News","volume":"32","author":"Zellers","year":"2019","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_3","unstructured":"Webber, B., Cohn, T., He, Y., and Liu, Y. (2020, January 16\u201320). Authorship Attribution for Neural Text Generation. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), Online."},{"key":"ref_4","unstructured":"Solaiman, I., Brundage, M., Clark, J., Askell, A., Herbert-Voss, A., Wu, J., Radford, A., Krueger, G., Kim, J.W., and Kreps, S. (2019). Release Strategies and the Social Impacts of Language Models. arXiv."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Gehrmann, S., Strobelt, H., and Rush, A.M. (2019). GLTR: Statistical Detection and Visualization of Generated Text. arXiv.","DOI":"10.18653\/v1\/P19-3019"},{"key":"ref_6","unstructured":"Mitchell, E., Lee, Y., Khazatsky, A., Manning, C.D., and Finn, C. (2023, January 23\u201329). DetectGPT: Zero-shot machine-generated text detection using probability curvature. Proceedings of the International Conference on Machine Learning, PMLR, Honolulu, HI, USA."},{"key":"ref_7","first-page":"17061","article-title":"A Watermark for Large Language Models","volume":"Volume 202","author":"Krause","year":"2023","journal-title":"Proceedings of the 40th International Conference on Machine Learning, PMLR"},{"key":"ref_8","unstructured":"Atallah, M.J., Raskin, V., Crogan, M., Hempelmann, C., Kerschbaum, F., Mohamed, D., and Naik, S. (2001, January 25\u201327). Natural language watermarking: Design, analysis, and a proof-of-concept implementation. Proceedings of the Information Hiding: 4th International Workshop, IH 2001, Pittsburgh, PA, USA. Proceedings 4."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Poje, K., Brcic, M., Kovac, M., and Babac, M.B. (2024). Effect of Private Deliberation: Deception of Large Language Models in Game Play. Entropy, 26.","DOI":"10.3390\/e26060524"},{"key":"ref_10","unstructured":"He, H., Liu, Y., Wang, Z., Mao, Y., and Bu, Y. (2024). Theoretically Grounded Framework for LLM Watermarking: A Distribution-Adaptive Approach. arXiv."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Fang, T., Jaggi, M., and Argyraki, K. (2017). Generating Steganographic Text with LSTMs. arXiv.","DOI":"10.18653\/v1\/P17-3017"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"65981","DOI":"10.1109\/ACCESS.2018.2866063","article-title":"AITSteg: An innovative text steganography technique for hidden transmission of text message via social media","volume":"6","author":"Ahvanooey","year":"2018","journal-title":"IEEE Access"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Iqbal, M.M., Khadam, U., Han, K.J., Han, J., and Jabbar, S. (2019, January 24\u201328). A robust digital watermarking algorithm for text document copyright protection based on feature coding. Proceedings of the 2019 15th InternationalWireless Communications & Mobile Computing Conference (IWCMC), Tangier, Morocco.","DOI":"10.1109\/IWCMC.2019.8766644"},{"key":"ref_14","unstructured":"Liu, Y., and Bu, Y. (2024). Adaptive Text Watermark for Large Language Models. arXiv."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Wang, Z., Gu, T., Wu, B., and Yang, Y. (2025). MorphMark: Flexible Adaptive Watermarking for Large Language Models. arXiv.","DOI":"10.3390\/math13091420"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"379","DOI":"10.1002\/j.1538-7305.1948.tb01338.x","article-title":"A mathematical theory of communication","volume":"27","author":"Shannon","year":"1948","journal-title":"Bell Syst. Tech. J."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"50","DOI":"10.1002\/j.1538-7305.1951.tb01366.x","article-title":"Prediction and entropy of printed English","volume":"30","author":"Shannon","year":"1951","journal-title":"Bell Syst. Tech. J."},{"key":"ref_18","unstructured":"Brown, T.B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., and Askell, A. (2020). Language Models are Few-Shot Learners. arXiv."},{"key":"ref_19","unstructured":"Cover, T.M., and Thomas, J.A. (2006). Elements of Information Theory, John Wiley & Sons. [2nd ed.]. Second Edition of the 1991 Original."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"1319","DOI":"10.1109\/18.669425","article-title":"Nonparametric entropy estimation for stationary processes and random fields, with applications to English text","volume":"44","author":"Kontoyiannis","year":"1998","journal-title":"IEEE Trans. Inf. Theory"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Bentz, C., Alikaniotis, D., Cysouw, M., and Ferrer-i Cancho, R. (2017). The entropy of words\u2014Learnability and expressivity across more than 1000 languages. Entropy, 19.","DOI":"10.20944\/preprints201704.0180.v1"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Bender, E.M., Gebru, T., McMillan-Major, A., and Shmitchell, S. (2021, January 3\u201310). On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?. Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency (FAccT \u201921), New York, NY, USA. FAccT \u201921.","DOI":"10.1145\/3442188.3445922"},{"key":"ref_23","first-page":"1","article-title":"PaLM: Scaling Language Modeling with Pathways","volume":"24","author":"Chowdhery","year":"2023","journal-title":"J. Mach. Learn. Res."},{"key":"ref_24","first-page":"27469","article-title":"Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defense","volume":"Volume 36","author":"Naumann","year":"2023","journal-title":"Proceedings of the Advances in Neural Information Processing Systems"},{"key":"ref_25","unstructured":"Sadasivan, V.S., Kumar, A., Balasubramanian, S., Wang, W., and Feizi, S. (2023). Can AI-Generated Text be Reliably Detected?. arXiv."},{"key":"ref_26","unstructured":"Jiang, Y., Wu, C., Boroujeny, M.K., Mark, B., and Zeng, K. (2025). StealthInk: A Multi-bit and Stealthy Watermark for Large Language Models. arXiv."},{"key":"ref_27","unstructured":"Diaa, A., Aremu, T., and Lukas, N. (2024). Optimized Adaptive Attacks against LLM Watermarks. arXiv."},{"key":"ref_28","unstructured":"Zhang, H., Edelman, B.L., Francati, D., Venturi, D., Ateniese, G., and Barak, B. (2023). Watermarks in the Sand: Impossibility of Strong Watermarking for Language Models. arXiv."},{"key":"ref_29","unstructured":"Christ, M., Gunn, S., and Zamir, O. (July, January 30). Undetectable Watermarks for Language Models. Proceedings of the The Thirty Seventh Annual Conference on Learning Theory, PMLR, Edmonton, AB, Canada."},{"key":"ref_30","unstructured":"Zhao, X., Wang, Y.X., and Li, L. (2023, January 1\u20135). Provable Robust Watermarking for AI-Generated Text. Proceedings of the International Conference on Learning Representations, Kigali, Rwanda."},{"key":"ref_31","unstructured":"Kuditipudi, R., Thickstun, J., Hashimoto, T., and Liang, P. (2023, January 23\u201329). Robust Distortion-free Watermarks for Language Models. Proceedings of the 40th International Conference on Machine Learning, Honolulu, HI, USA."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Fernandez, P., Chaffin, A., Tit, K., Chizat, V., and Furon, T. (2023, January 4\u20137). Three Bricks to Consolidate Watermarks for Large Language Models. Proceedings of the 2023 IEEE International Workshop on Information Forensics and Security (WIFS), Nurnberg, Germany.","DOI":"10.1109\/WIFS58808.2023.10374576"},{"key":"ref_33","unstructured":"Piet, J., Sitawarin, C., Fang, V., Mu, N., and Wagner, D. (2023). Mark My Words: Analyzing and Evaluating Language Model Watermarks. arXiv."},{"key":"ref_34","unstructured":"Wu, Y., Hu, Z., Guo, J., Zhang, H., and Huang, H. (2023). A Resilient and Accessible Distribution-Preserving Watermark for Large Language Models. arXiv."},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Hou, A.B., Zhang, J., He, T., Wang, Y., Chuang, Y.S., Wang, H., Shen, L., Van Durme, B., Khashabi, D., and Tsvetkov, Y. (2023). SemStamp: A Semantic Watermark with Paraphrastic Robustness for Text Generation. arXiv.","DOI":"10.18653\/v1\/2024.naacl-long.226"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Pennington, J., Socher, R., and Manning, C.D. (2014, January 25\u201329). Glove: Global vectors for word representation. Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), Doha, Qatar.","DOI":"10.3115\/v1\/D14-1162"},{"key":"ref_37","unstructured":"Heafield, K. (2011, January 30\u201331). KenLM: Faster and Smaller Language Model Queries. Proceedings of the Sixth Workshop on Statistical Machine Translation, Edinburgh, UK."},{"key":"ref_38","first-page":"9","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford","year":"2019","journal-title":"OpenAI Blog"},{"key":"ref_39","unstructured":"Chiang, W.L., Li, Z., Lin, Z., Sheng, Y., Wu, Z., Zhang, H., Zheng, L., Zhuang, S., Zhuang, Y., and Gonzalez, J.E. (2025, May 14). Vicuna: An Open-Source Chatbot Impressing GPT-4 with 90%* ChatGPT Quality. Available online: https:\/\/lmsys.org\/blog\/2023-03-30-vicuna\/."},{"key":"ref_40","unstructured":"Merity, S., Xiong, C., Bradbury, J., and Socher, R. (2016). Pointer Sentinel Mixture Models. arXiv."},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Narayan, S., Cohen, S.B., and Lapata, M. (2018). Don\u2019t give me the details, just the summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization. arXiv.","DOI":"10.18653\/v1\/D18-1206"},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Lewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., Stoyanov, V., and Zettlemoyer, L. (2019). BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension. arXiv.","DOI":"10.18653\/v1\/2020.acl-main.703"},{"key":"ref_43","first-page":"1","article-title":"Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer","volume":"21","author":"Raffel","year":"2020","journal-title":"J. Mach. Learn. Res."},{"key":"ref_44","unstructured":"Jiang, A.Q., Sablayrolles, A., Mensch, A., Bamford, C., Chaplot, D.S., de las Casas, D., Bressand, F., Lengyel, G., Lample, G., and Saulnier, L. (2023). Mistral 7B. arXiv."},{"key":"ref_45","unstructured":"Shi, Y., Yang, Y., Sheng, Q., Mi, H., Hu, B., Xu, C., and Cao, J. (2025). PhantomHunter: Detecting Unseen Privately-Tuned LLM-Generated Text via Family-Aware Learning. arXiv."},{"key":"ref_46","unstructured":"Bhattacharjee, A., Moraffah, R., Garland, J., and Liu, H. (2024). EAGLE: A Domain Generalization Framework for AI-generated Text Detection. arXiv."},{"key":"ref_47","unstructured":"Xu, Y., Wang, Y., Bi, Y., Cao, H., Lin, Z., Zhao, Y., and Wu, F. (2024). Training-free LLM-generated Text Detection by Mining Token Probability Sequences. arXiv."}],"container-title":["Entropy"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1099-4300\/27\/8\/784\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T18:15:26Z","timestamp":1760033726000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1099-4300\/27\/8\/784"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,24]]},"references-count":47,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2025,8]]}},"alternative-id":["e27080784"],"URL":"https:\/\/doi.org\/10.3390\/e27080784","relation":{},"ISSN":["1099-4300"],"issn-type":[{"value":"1099-4300","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,24]]}}}