Elamur Research⌕
☰
home / papers / 2026 / FAID: Fine-grained AI-generated Text Detection using Multi-task Auxiliary and Multi-level Contrastive Learning

FAID: Fine-grained AI-generated Text Detection using Multi-task Auxiliary and Multi-level Contrastive Learning

Paper · 2026 · Minh Ngoc Ta, Dong Cao Van, Duc-Anh Hoang et al.

Summary

Detecting machine-written text is harder when humans and LLMs co-write, and most detectors only output a human vs. AI label or work only in English. The authors frame detection as a three-way task: fully human, fully LLM, or human–LLM collaborative. They also try to identify which LLM family produced the text. They release FAIDSet, an academic-domain corpus in English and Vietnamese. Its texts come from student theses and paper abstracts, with machine and collaborative versions produced by several model families. Their detector, FAID, trains an encoder with two kinds of objective. A multi-level contrastive objective treats each LLM family as a separate stylistic "author", and auxiliary classification tasks are added alongside it. At inference, FAID labels a text by finding similar texts in a stored embedding database with Fuzzy k-Nearest Neighbors. New domains or generators can be handled by adding their embeddings to this database, with no retraining. FAID beats several baselines, with the largest gains on unseen domains and generators. This makes it a promising tool for transparent disclosure of AI assistance in multilingual academic writing.

Key points

  • FAIDSet has 83,350 examples in English and Vietnamese.
    • Theses come from Hanoi University of Science and Technology; abstracts come from arXiv and VJOL.
    • It is available on Hugging Face (ngocminhta/FAIDSet), and the code is at github.com/mbzuai-nlp/FAID.
  • Generators and collaboration types:
    • The generator families are GPT-4/4o, Llama-3.x, Gemini 2.x and DeepSeek V3/R1.
    • Collaborative texts are LLM-polished, LLM-continued or LLM-paraphrased versions of human text.
    • Prompts varied persona and tone to diversify outputs.
  • Method: multi-level contrastive learning is combined with multi-task auxiliary classification. Inference uses Fuzzy k-NN retrieval, and adaptation to new data is incremental and training-free.
  • In-the-wild test: five volunteers co-wrote 200 texts with ChatGPT, Gemini, DeepSeek and Llama 3.1. FAID reached 88.5% accuracy, 85.9% precision and 89.7% recall on these texts.
  • Limitations:
    • The data is synthetic and covers few low-resource languages.
    • Attribution becomes unreliable when one text has been shaped by several LLMs.
    • The authors say FAID should not be used for high-stakes decisions without human oversight.

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

🎓Mohamed bin Zayed University of Artificial IntelligenceAbu Dhabi · United Arab Emirates124 works · 5,676 citations🎓Sofia University "St. Kliment Ohridski"Sofia · Bulgaria21 works · 2,595 citations🎓Hanoi University of Science and TechnologyHanoi · Vietnam10 works · 89 citations🎓Trường Đại học Khoa học và Công nghệ Hà NộiHanoi · Vietnam8 works · 88 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 works

Cite

Cite as Minh Ngoc Ta, Dong Cao Van, Duc-Anh Hoang, Minh Le-Anh, Truong Nguyen, My Anh Tran Nguyen, Yuxia Wang, Preslav Nakov and Dinh Viet Sang. 2026. FAID: Fine-grained AI-generated Text Detection using Multi-task Auxiliary and Multi-level Contrastive Learning. Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.eacl-long.151

@inproceedings{ta2026faid,
  title = {{FAID: Fine-grained AI-generated Text Detection using Multi-task Auxiliary and Multi-level Contrastive Learning}},
  author = {Minh Ngoc Ta, Dong Cao Van, Duc-Anh Hoang, Minh Le-Anh, Truong Nguyen, My Anh Tran Nguyen, Yuxia Wang, Preslav Nakov and Dinh Viet Sang},
  booktitle = {Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.eacl-long.151},
  url = {https://doi.org/10.18653/v1/2026.eacl-long.151}
}

References

39 references, 12 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Mervat Abassy, Kareem Elozeiri, Alexander Aziz, Minh Ngoc Ta, Raj Vardhan Tomar, Bimarsha Adhikari, Saad El Dine Ahmed, Yuxia Wang, Osama Mohammed Afzal, Zhuohan Xie, Jonibek Mansurov, Ekaterina Artemova, Vladislav Mikhailov, Rui Xing, Jiahui Geng, Hasan Iqbal, Zain Muhammad Mujahid, Tarek Mahmoud,… find
  2. Ekaterina Artemova, Jason S Lucas, Saranya Venkatraman, Jooyoung Lee, Sergei Tilga, Adaku Uchendu, and Vladislav Mikhailov. 2025. Beemo: Benchmark of expert-edited machine-generated outputs. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computati… find
  3. Guangsheng Bao, Lihua Rong, Yanbin Zhao, Qiji Zhou, and Yue Zhang. 2025. Decoupling content and expression: Two-dimensional detection of AI-generated text. ArXiv preprint, arXiv:2503.00258. — [arXiv:2503.00258](https://arxiv.org/abs/2503.00258) find
  4. Rich Caruana. 1997. Multitask learning. Machine learning, 28:41–75. find
  5. Tuhin Chakrabarty, Philippe Laban, and Chien-Sheng Wu. 2025. Can AI writing be salvaged? Mitigating idiosyncrasies and improving human-AI alignment in the writing process through edits. In Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems, CHI ’25, Yokohama, Japan. Associa… find
  6. Ting Chen, Simon Kornblith, Mohammad Norouzi, and Geoffrey Hinton. 2020. A simple framework for contrastive learning of visual representations. In Proceedings of the 37th International Conference on Machine Learning, ICML ’20, pages 1597–1607. JMLR.org.
  7. Yutian Chen, Hao Kang, Vivian Zhai, Liangze Li, Rita Singh, and Bhiksha Raj. 2023. Token prediction as implicit classification to identify LLM-generated text. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, EMNLP ’2023, pages 13112–13120, Singapore. Associa… find
  8. Zihao Cheng, Li Zhou, Feng Jiang, Benyou Wang, and Haizhou Li. 2025. Beyond binary: Towards fine-grained LLM-generated text detection via role recognition and involvement measurement. In Proceedings of the ACM on Web Conference 2025, WWW ’25, page 2677–2688. ACM. find
  9. Hyung Won Chung, Le Hou, Shayne Longpre, Barret Zoph, Yi Tay, William Fedus, Yunxuan Li, Xuezhi Wang, Mostafa Dehghani, Siddhartha Brahma, Albert Webson, Shixiang Shane Gu, Zhuyun Dai, Mirac Suzgun, Xinyun Chen, Aakanksha Chowdhery, Alex Castro-Ros, Marie Pellat, Kevin Robinson, Dasha Valter, Sharan… find
  10. Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer, and Veselin Stoyanov. 2020. Unsupervised cross-lingual representation learning at scale. In Proceedings of the 58th Annual Meeting of the Association fo… find
  11. DeepSeek-AI. 2025. DeepSeek-V3 technical report. ArXiv preprint, arXiv:2412.19437. find
  12. Matthijs Douze, Alexandr Guzhva, Chengqi Deng, Jeff Johnson, Gergely Szilvasy, Pierre-Emmanuel Mazaré, Maria Lomeli, Lucas Hosseini, and Hervé Jégou. 2025. The Faiss library. ArXiv preprint, arXiv:2401.08281.
  13. Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, et al. 2024. The Llama 3 herd of models. ArXiv preprint, arXiv:2407.21783.
  14. Liam Dugan, Daphne Ippolito, Arun Kirubarajan, Sherry Shi, and Chris Callison-Burch. 2022. Real or fake text?: Investigating human ability to detect boundaries between human-written and machine-generated text. ArXiv preprint, arXiv:2212.12672. — [arXiv:2212.12672](https://arxiv.org/abs/2212.12672) find
  15. Tianyu Gao, Xingcheng Yao, and Danqi Chen. 2021. SimCSE: Simple contrastive learning of sentence embeddings. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, EMNLP ’2021, pages 6894–6910, Online and Punta Cana, Dominican Republic. Association for Computation…
  16. Gemini Team. 2024. Gemini: A family of highly capable multimodal models. ArXiv preprint, arXiv:2312.11805.
  17. John Giorgi, Osvald Nitski, Bo Wang, and Gary Bader. 2021. DeCLUTR: Deep contrastive learning for unsupervised textual representations. ArXiv preprint, arXiv:2006.03659.
  18. Xun Guo, Shan Zhang, Yongxin He, Ting Zhang, Wanquan Feng, Haibin Huang, and Chongyang Ma. 2024. DeTeCtive: Detecting AI-generated text via multi-level contrastive learning. In Advances in Neural Information Processing Systems, volume 37 of NeurIPS ’2024, pages 88320–88347. Curran Associates, Inc. find
  19. James Hutson. 2025. Human-AI collaboration in writing: A multidimensional framework for creative and intellectual authorship. International Journal of Changes in Education. find
  20. Ryuto Koike, Masahiro Kaneko, and Naoaki Okazaki. 2024. OUTFOX: LLM-generated essay detection through in-context learning with adversarially generated examples. In Proceedings of the 38th AAAI Conference on Artificial Intelligence, AAAI ’2024, pages 21258–21266, Vancouver, Canada. find
  21. Laida Kushnareva, Tatiana Gaintseva, German Magai, Serguei Barannikov, Dmitry Abulkhanov, Kristian Kuznetsov, Eduard Tulchinskii, Irina Piontkovskaya, and Sergey Nikolenko. 2024. AI-generated text boundary detection with RoFT. ArXiv preprint, arXiv:2311.08349. — [arXiv:2311.08349](https://arxiv.org/abs/2311.08349) find
  22. Mina Lee, Percy Liang, and Qian Yang. 2022. CoAuthor: Designing a human-AI collaborative writing dataset for exploring language model capabilities. In Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems, CHI ’22, New Orleans, LA, USA. Association for Computing Machinery. find
  23. Yafu Li, Qintong Li, Leyang Cui, Wei Bi, Zhilin Wang, Longyue Wang, Linyi Yang, Shuming Shi, and Yue Zhang. 2024. MAGE: Machine-generated text detection in the wild. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL ’2024, pages 3…
  24. Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. RoBERTa: A robustly optimized BERT pretraining approach. ArXiv preprint, arXiv:1907.11692.
  25. Dominik Macko, Robert Moro, and Ivan Srba. 2025. Increasing the robustness of the fine-tuned multilingual machine-generated text detectors. ArXiv preprint, arXiv:2503.15128. — [arXiv:2503.15128](https://arxiv.org/abs/2503.15128) find
  26. OpenAI. 2024. GPT-4o system card. ArXiv preprint, arXiv:2410.21276.
  27. Shoumik Saha and Soheil Feizi. 2025. Almost AI, almost human: The challenge of detecting AI-polished writing. ArXiv preprint, arXiv:2502.15666. find
  28. Jinyan Su, Terry Zhuo, Di Wang, and Preslav Nakov. 2023. DetectLLM: Leveraging log rank information for zero-shot detection of machine-generated text. In Findings of the Association for Computational Linguistics, EMNLP ’2023, pages 12395–12412, Singapore. Association for Computational Linguistics.
  29. Jannis Vamvas and Rico Sennrich. 2023. Towards unsupervised recognition of token-level semantic differences in related documents. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, EMNLP ’2023, Singapore. Association for Computational Linguistics. find
  30. Liang Wang, Nan Yang, Xiaolong Huang, Binxing Jiao, Linjun Yang, Daxin Jiang, Rangan Majumder, and Furu Wei. 2024a. Text embeddings by weakly-supervised contrastive pre-training. ArXiv preprint, arXiv:2212.03533.
  31. Liang Wang, Nan Yang, Xiaolong Huang, Linjun Yang, Rangan Majumder, and Furu Wei. 2024b. Multilingual E5 text embeddings: A technical report. ArXiv preprint, arXiv:2402.05672.
  32. Pengyu Wang, Linyang Li, Ke Ren, Botian Jiang, Dong Zhang, and Xipeng Qiu. 2023. SeqXGPT: Sentence-level AI-generated text detection. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, EMNLP ’2023, pages 1144–1156, Singapore. Association for Computational Ling… find
  33. Yuxia Wang, Jonibek Mansurov, Petar Ivanov, Jinyan Su, Artem Shelmanov, Akim Tsvigun, Osama Mohammed Afzal, Tarek Mahmoud, Giovanni Puccetti, and Thomas Arnold. 2024c. SemEval-2024 task 8: Multidomain, multimodel and multilingual machine-generated text detection. In Proceedings of the 18th Internati… find
  34. Yuxia Wang, Jonibek Mansurov, Petar Ivanov, Jinyan Su, Artem Shelmanov, Akim Tsvigun, Osama Mohammed Afzal, Tarek Mahmoud, Giovanni Puccetti, Thomas Arnold, Alham Aji, Nizar Habash, Iryna Gurevych, and Preslav Nakov. 2024d. M4GT-Bench: Evaluation benchmark for black-box machine-generated text detect… find
  35. Yuxia Wang, Jonibek Mansurov, Petar Ivanov, Jinyan Su, Artem Shelmanov, Akim Tsvigun, Chenxi Whitehouse, Osama Mohammed Afzal, Tarek Mahmoud, Toru Sasaki, Thomas Arnold, Alham Fikri Aji, Nizar Habash, Iryna Gurevych, and Preslav Nakov. 2024e. M4: Multi-generator, multi-domain, and multi-lingual blac… find
  36. Yuxia Wang, Artem Shelmanov, Jonibek Mansurov, Akim Tsvigun, Vladislav Mikhailov, Rui Xing, Zhuohan Xie, Jiahui Geng, Giovanni Puccetti, Ekaterina Artemova, Jinyan Su, Minh Ngoc Ta, Mervat Abassy, Kareem Ashraf Elozeiri, Saad El Dine Ahmed El Etter, Maiya Goloburda, Tarek Mahmoud, Raj Vardhan Tomar,… find
  37. Zijie Zeng, Shiqi Liu, Lele Sha, Zhuang Li, Kaixun Yang, Sannyuya Liu, Dragan Gašević, and Guanliang Chen. 2024. Detecting AI-generated sentences in human-AI collaborative hybrid texts: Challenges, strategies, and insights. In Proceedings of the Thirty-Third International Joint Conference on Artifi… find
  38. Zijie Zeng, Lele Sha, Yuheng Li, Kaixun Yang, Dragan Gašević, and Guanliang Chen. 2023. Towards automatic boundary detection for human-AI collaborative hybrid essay in education. ArXiv preprint, arXiv:2307.12267. — [arXiv:2307.12267](https://arxiv.org/abs/2307.12267) find
  39. Qihui Zhang, Chujie Gao, Dongping Chen, Yue Huang, Yixin Huang, Zhenyang Sun, Shilin Zhang, Weiye Li, Zhengyan Fu, Yao Wan, and Lichao Sun. 2024. LLM-as-a-Coauthor: Can mixed human-written and machine-generated text be detected? In Findings of the Association for Computational Linguistics, NAACL ’20… find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/ta-2026-faid-fine-grained-ai-generated/