Elamur Research⌕
☰
home / papers / 2026 / Measuring Human Contribution in AI-Assisted Content Generation

Measuring Human Contribution in AI-Assisted Content Generation

Paper · 2026 · Yueqi Xie, Tao Qi, Jingwei Yi et al.

Summary

As more writing is produced by LLMs following human prompts, deciding how much of a piece reflects the person rather than the model has become difficult. This matters for education, research integrity and copyright. The authors frame this as a measurement problem and propose an information-theoretic score. It is the share of the output's self-information, I(y) = −log p(y), that is explained by the human input x. In other words, the score is the mutual information I(x; y) = I(y) − I(y|x) divided by I(y), which usually yields a value between 0 and 1.

To test it, they built a dataset with known contribution levels in four domains. They prompted several LLMs with inputs ranging from rich to sparse: a full text to polish, a summary, a title, or a subject. The score reliably separated these levels across domains and models. It also held up when the authors varied length, style and temperature, and when they applied two adaptive attacks.

An extension estimates a lower bound on the minimum human contribution from the text alone, with no access to the prompt. This makes the approach more practical for real-world use.

Key points

  • Dataset: 2,000 sampled entries per domain from four corpora:
    • paper abstracts (Arxiv),
    • news (News Articles),
    • patent abstracts (HUPD),
    • poems (Poetry Foundation).
  • Generation setup: Missing summaries, titles and subjects were filled in with GPT-3.5. Outputs came from Llama-3, Mixtral, GPT-3.5, GPT-4o, Gemini-2.0 Flash and Claude-3.5 Haiku.
  • Ordering of contribution levels: The measure ranks the levels as expected, with polishing highest, then summary, then title and subject. The authors also applied it to real human–AI co-creation data.
  • Input-free estimator: It replaces I(y|x) with an empirically chosen bound, so some estimates can fall below zero. Its relative rankings matched human annotations in 92.87% of cases.
  • Limitations:
    • The method covers text only.
    • It measures informational contribution, which is not the same as creativity or authorship.
    • It does not settle copyright questions.

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

🎓Tsinghua UniversityBeijing · China333 works · 87,019 citations🎓University of Illinois Urbana-ChampaignUrbana · United States184 works · 69,372 citations🎓Peking UniversityBeijing · China244 works · 39,359 citations🎓Princeton UniversityPrinceton · United States90 works · 39,109 citations🏢Microsoft Research AsiaBeijing · China73 works · 60,539 citations🎓University of Science and Technology of ChinaHefei · China108 works · 112,698 citations🎓University of Hong KongPok Fu Lam · China89 works · 11,774 citations

 Fields, topics and concepts

📚 Shelves 7The 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 worksSocial Sciences · 1,772 worksSafety Research · 543 worksGeneral Social Sciences · 227 works

Cite

Cite as Yueqi Xie, Tao Qi, Jingwei Yi, Xiyuan Yang, Ryan Whalen, Junming Huang, Qian Ding, Yu Xie, Xing Xie and Fangzhao Wu. 2026. Measuring Human Contribution in AI-Assisted Content Generation. Proceedings of the 64th Annual Meeting 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.acl-long.279

@inproceedings{xie2026measuring,
  title = {{Measuring Human Contribution in AI-Assisted Content Generation}},
  author = {Yueqi Xie, Tao Qi, Jingwei Yi, Xiyuan Yang, Ryan Whalen, Junming Huang, Qian Ding, Yu Xie, Xing Xie and Fangzhao Wu},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.279},
  url = {https://doi.org/10.18653/v1/2026.acl-long.279}
}

References

50 references, 11 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Ryan Abbott and Elizabeth Rothman. 2023. Disrupting creativity: Copyright law in the age of generative artificial intelligence. Fla. L. Rev., 75:1141. find
  2. Anthropic. 2023. Model card and evaluations for claude models. find
  3. Saeed Awadh Bin-Nashwan, Mouad Sadallah, and Mohamed Bouteraa. 2023. Use of chatgpt in academia: Academic integrity hangs in the balance. Technology in Society, 75:102370. find
  4. Wei-Lin Chiang, Lianmin Zheng, Ying Sheng, Anastasios Nikolas Angelopoulos, Tianle Li, Dacheng Li, Hao Zhang, Banghua Zhu, Michael Jordan, Joseph E Gonzalez, and 1 others. 2024. Chatbot arena: An open platform for evaluating llms by human preference. arXiv preprint arXiv:2403.04132.
  5. Kevin Clark, Deepak Narayanan, Jiseung Kwon, Aaron Harlap, Matei Zaharia, and Christopher Re. 2023. Mixtral of experts. arXiv preprint arXiv:2401.04088. find
  6. Ismail Dergaa, Karim Chamari, Piotr Zmijewski, and Helmi Ben Saad. 2023. From human writing to artificial intelligence generated text: examining the prospects and potential threats of chatgpt in academic writing. Biology of Sport, 40(2):615–622. find
  7. Fiona Fui-Hoon Nah, Ruilin Zheng, Jingyuan Cai, Keng Siau, and Langtao Chen. 2023. Generative ai and chatgpt: Applications, challenges, and ai-human collaboration. find
  8. Tiana Garbett and James Gatto. 2023. Generative ai and copyright – some recent denials and unanswered questions. https://www.intellectualproperty lawblog.com/archives/generative-ai-and-c opyright-some-recent-denials-and-unanswe red-questions/. ↗
  9. Biyang Guo, Xin Zhang, Ziyuan Wang, Minqi Jiang, Jinran Nie, Yuxuan Ding, Jianwei Yue, and Yupeng Wu. 2023. How close is chatgpt to human experts? comparison corpus, evaluation, and detection. arXiv preprint arXiv:2301.07597.
  10. Abhimanyu Hans, Avi Schwarzschild, Valeriia Cherepanova, Hamid Kazemi, Aniruddha Saha, Micah Goldblum, Jonas Geiping, and Tom Goldstein. 2024. Spotting llms with binoculars: Zero-shot detection of machine-generated text. arXiv preprint arXiv:2401.12070. find
  11. Wei Hao, Ran Li, Weiliang Zhao, Junfeng Yang, and Chengzhi Mao. 2025. Learning to rewrite: Generalized LLM-generated text detection. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 6421–6434, Vienna, Austria. Association for C… find
  12. Patrick Hemmer, Monika Westphal, Max Schemmer, Sebastian Vetter, Michael Vössing, and Gerhard Satzger. 2023. Human-ai collaboration: The effect of ai delegation on human task performance and task satisfaction. In Proceedings of the 28th International Conference on Intelligent User Interfaces, IUI ’2… find
  13. Abe Bohan Hou, Jingyu Zhang, Tianxing He, Yichen Wang, Yung-Sung Chuang, Hongwei Wang, Lingfeng Shen, Benjamin Van Durme, Daniel Khashabi, and Yulia Tsvetkov. 2023. Semstamp: A semantic watermark with paraphrastic robustness for text generation. arXiv preprint arXiv:2310.03991. find
  14. Kalin Hristov. 2016. Artificial intelligence and the copyright dilemma. Idea, 57:431. find
  15. James Hutson. 2024. Rethinking plagiarism in the era of generative ai. Journal of Intelligent Communication, 4(1):20–31. find
  16. John Kirchenbauer, Jonas Geiping, Yuxin Wen, Jonathan Katz, Ian Miers, and Tom Goldstein. 2023. A watermark for large language models. arXiv preprint arXiv:2301.10226.
  17. Sachin Kumar, Vidhisha Balachandran, Lucille Njoo, Antonios Anastasopoulos, and Yulia Tsvetkov. 2023. Language generation models can cause harm: So what can we do about it? an actionable survey. In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Lingui…
  18. Diana Kwon. 2024. Ai is complicating plagiarism. how should scientists respond? Nature. find
  19. Thomas Lavergne, Tanguy Urvoy, and François Yvon. 2008. Detecting fake content with relative entropy scoring. PAN, 8:27–31. find
  20. Microsoft. 2023. Introducing microsoft 365 copilot – your copilot for work. https://blogs.microsoft.com/blog/2023/03/16/introducing-microsoft -365-copilot-your-copilot-for-work/. ↗
  21. Eric Mitchell, Yoonho Lee, Alexander Khazatsky, Christopher D Manning, and Chelsea Finn. 2023. Detectgpt: Zero-shot machine-generated text detection using probability curvature. arXiv preprint arXiv:2301.11305.
  22. Ryosuke Nakadai, Yo Nakawake, and Shota Shibasaki. 2023. Ai language tools risk scientific diversity and innovation. Nature Human Behaviour, 7(11):1804– 1805. find
  23. Voice of America. 2023. Schools ban chatgpt amid fears of artificial intelligence-assisted cheating. VOA News. Accessed: 2024-06-27. find
  24. OpenAI. 2022. Chatgpt. https://openai.com/blog/ chatgpt. ↗
  25. OpenAI. 2023. AI text classifier. find
  26. Yikang Pan, Liangming Pan, Wenhu Chen, Preslav Nakov, Min-Yen Kan, and William Yang Wang. 2023. On the risk of misinformation pollution with large language models. arXiv preprint arXiv:2305.13661.
  27. Claude Elwood Shannon. 1948. A mathematical theory of communication. The Bell system technical journal, 27(3):379–423. find
  28. Natasha Singer. 2023. Ban or embrace? colleges wrestle with a.i.-generated admissions essays. The New York Times. Accessed: 2024-06-27. find
  29. Gemini Team, Rohan Anil, Sebastian Borgeaud, Jean-Baptiste Alayrac, Jiahui Yu, Radu Soricut, Johan Schalkwyk, Andrew M Dai, Anja Hauth, Katie Millican, and 1 others. 2023. Gemini: a family of highly capable multimodal models. arXiv preprint arXiv:2312.11805.
  30. Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, and 1 others. 2023. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971.
  31. U.S. Copyright Office. 2023a. Cancellation decision re: Zarya of the dawn (vau001480196). At 2. find
  32. U.S. Copyright Office. 2023b. Copyright registration guidance: Works containing material generated by artificial intelligence. find
  33. U.S. Copyright Office Review Board. 2022. Decision affirming refusal of registration of a recent entrance to paradise. At 2. find
  34. Valentina Pyatkin. 2024. Prompt classification. find
  35. Vivek Verma, Eve Fleisig, Nicholas Tomlin, and Dan Klein. 2024. Ghostbuster: Detecting text ghostwritten by large language models. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)…
  36. Dakuo Wang, Elizabeth Churchill, Pattie Maes, Xiangmin Fan, Ben Shneiderman, Yuanchun Shi, and Qianying Wang. 2020. From human-human collaboration to human-ai collaboration: Designing ai systems that can work together with people. In Extended abstracts of the 2020 CHI conference on human factors in… find
  37. Roosa Wingström, Johanna Hautala, and Riina Lundman. 2024. Redefining creativity in the era of ai? perspectives of computer scientists and new media artists. Creativity Research Journal, 36(2):177–193. find
  38. Junchao Wu, Shu Yang, Runzhe Zhan, Yulin Yuan, Lidia Sam Chao, and Derek Fai Wong. 2025. A survey on llm-generated text detection: Necessity, methods, and future directions. Computational Linguistics, 51(1):275–338.
  39. Yueqi Xie, Jingwei Yi, Jiawei Shao, Justin Curl, Lingjuan Lyu, Qifeng Chen, Xing Xie, and Fangzhao Wu. 2023. Defending chatgpt against jailbreak attack via self-reminders. Nature Machine Intelligence, 5(12):1486–1496. find
  40. Xianjun Yang, Wei Cheng, Linda Petzold, William Yang Wang, and Haifeng Chen. 2023a. Dna-gpt: Divergent n-gram analysis for training-free detection of gpt-generated text. arXiv preprint arXiv:2305.17359. — [arXiv:2305.17359](https://arxiv.org/abs/2305.17359) find
  41. Xianjun Yang, Liangming Pan, Xuandong Zhao, Haifeng Chen, Linda Petzold, William Yang Wang, and Wei Cheng. 2023b. A survey on detection of llms-generated content. arXiv preprint arXiv:2310.15654. find
  42. Xianjun Yang, Xiao Wang, Qi Zhang, Linda Petzold, William Yang Wang, Xun Zhao, and Dahua Lin. 2023c. Shadow alignment: The ease of subverting safely-aligned language models. arXiv preprint arXiv:2310.02949. find
  43. Hao Yu. 2023. Reflection on whether chat gpt should be banned by academia from the perspective of education and teaching. Frontiers in Psychology, 14:1181712. find
  44. Peipeng Yu, Jiahan Chen, Xuan Feng, and Zhihua Xia. 2025. Cheat: A large-scale dataset for detecting chatgpt-written abstracts. IEEE Transactions on Big Data, 11(3):898–906. find
  45. Haolan Zhan, Xuanli He, Qiongkai Xu, Yuxiang Wu, and Pontus Stenetorp. 2023. G3detector: General gpt-generated text detector. arXiv preprint arXiv:2305.12680. — [arXiv:2305.12680](https://arxiv.org/abs/2305.12680) find
  46. Rongsheng Zhang, Xiaoxi Mao, Le Li, Lin Jiang, Lin Chen, Zhiwei Hu, Yadong Xi, Changjie Fan, and Minlie Huang. 2022. Youling: an ai-assisted lyrics creation system. arXiv preprint arXiv:2201.06724. find
  47. Wenting Zhao, Xiang Ren, Jack Hessel, Claire Cardie, Yejin Choi, and Yuntian Deng. 2024. Wildchat: 1m chatgpt interaction logs in the wild. arXiv preprint arXiv:2405.01470.
  48. Xuandong Zhao, Prabhanjan Vijendra Ananth, Lei Li, and Yu-Xiang Wang. 2023. Provable robust watermarking for ai-generated text. ArXiv, abs/2306.17439. find
  49. Chengbo Zheng, Dakuo Wang, April Yi Wang, and Xiaojuan Ma. 2022. Telling stories from computational notebooks: Ai-assisted presentation slides creation for presenting data science work. In Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems, pages 1–20. find
  50. Contents \#\# Introduction \#\# Related Work Dataset Construction Methods 4.1 Defining Human Contribution in AI-Assisted Generation . . . . . . 4.2 Estimating Human Contribution Without Human Input . . . . . . Experiments 5.1 Human Contribution Evaluation . 5.2 Impact of Writing Style . . . . . . 5.3 Ge… find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/xie-2026-measuring-human-contribution-ai-assisted/