pith. sign in

arxiv: 2606.04205 · v1 · pith:XFXBHUSCnew · submitted 2026-06-02 · 💻 cs.MM · cs.AI· cs.CL· cs.CV· cs.LG· cs.SD

DetectZoo: A Unified Toolkit for AI-Generated Content Detection Across Text, Audio, and Image Modalities

Sajad Ebrahimi , Nima Jamali , Bardia Shirsalimian , Kelly McConvey , Wentao Zhang , Jalehsadat Mahdavimoghaddam , Maksym Taranukhin , Maura Grossman
show 3 more authors
Vered Shwartz Yuntian Deng Ebrahim Bagheri
This is my paper

Pith reviewed 2026-06-28 07:05 UTC · model grok-4.3

classification 💻 cs.MM cs.AIcs.CLcs.CVcs.LGcs.SD
keywords AI-generated content detectionmultimodal detection toolkittext image audio forensicsbenchmarking frameworkreproducible evaluationunified detection APIgenerative model detection
0
0 comments X

The pith

DetectZoo supplies one interface that runs 61 detectors on 22 datasets for spotting machine-generated text, audio, and images.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces DetectZoo as a toolkit that places existing detectors and datasets behind a single API so the full pipeline from data loading through evaluation stays consistent. This addresses the fragmentation where each detector previously used its own code, preprocessing steps, and metrics, making direct comparisons hard. A sympathetic reader would see the value in being able to reproduce published results and run fair benchmarks across modalities without rewriting adapters. The work centers on making multi-modal detection research more practical and reproducible.

Core claim

DetectZoo provides reference implementations of 61 detectors, native loaders for 22 benchmark datasets, and a standardized evaluation pipeline that reports multiple metrics through a common interface; each detector remains self-contained, automatically caches pretrained weights, and reproduces the original published results.

What carries the argument

The unified API that standardizes the empirical pipeline from data ingestion and preprocessing to model assessment across text, audio, and image modalities.

If this is right

  • Researchers can now compare detectors systematically without custom integration code for each one.
  • Performance gaps between modalities become visible through the shared metrics.
  • New detectors can be added while keeping the same evaluation protocol.
  • Reproduction of results becomes automatic for the included baselines.
  • Development of cross-modal detection methods gains a common testbed.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Widespread use could push the field toward shared evaluation standards rather than bespoke ones.
  • The modular design makes it feasible to test whether a single detector family works across modalities without reimplementation.
  • Community additions of newer detectors would keep the toolkit current without breaking existing comparisons.

Load-bearing premise

Wrapping the original detectors into a shared interface leaves their behavior and published performance unchanged.

What would settle it

Running the toolkit versions of published detectors on their original test sets and obtaining different performance numbers than the papers reported.

Figures

Figures reproduced from arXiv: 2606.04205 by Bardia Shirsalimian, Ebrahim Bagheri, Jalehsadat Mahdavimoghaddam, Kelly McConvey, Maksym Taranukhin, Maura Grossman, Nima Jamali, Sajad Ebrahimi, Vered Shwartz, Wentao Zhang, Yuntian Deng.

Figure 1
Figure 1. Figure 1: Overview of DetectZoo. The framework standardizes the evaluation of AI-generated content detectors across text, image, and audio. It provides a unified API from data ingestion to metric computation. A growing body of research has responded by developing detection methods for AI-generated content across text [8, 9, 10], images [11, 12, 13], and audio [14, 15]. Each new method is typically re￾leased as a sta… view at source ↗
read the original abstract

The growing popularity and capacity of generative models have eroded the distinction between human and machine-generated content, motivating a growing body of work on detection across text, images, and audio. Most available detectors are either commercial software or, if open-source, come with incompatible codebases with bespoke preprocessing, evaluation protocols, and evaluation metrics, which make their adoption, fair comparison, and reproduction quite difficult. To address this critical gap, we introduce DetectZoo, a first-of-its-kind, extensible toolkit designed to provide a unified interface for AI-generated content detection across text, audio, and image modalities. DetectZoo standardizes the complete empirical pipeline, from data ingestion and preprocessing to model assessment, offering researchers a cohesive framework to benchmark state-of-the-art detectors systematically. By integrating diverse public datasets and baseline detection algorithms under a single, unified API, our toolkit facilitates rigorous and reproducible evaluation. DetectZoo provides reference implementations of 61 detectors, native loaders for 22 benchmark datasets, and a standardized evaluation pipeline that reports multiple metrics through a common interface. Each detector is self-contained yet accessible through the same interface, automatically caches pretrained weights, and reproduces the original published results. DetectZoo lowers the barrier to entry for multi-modal AI forensics, enabling researchers to identify performance gaps across domains and accelerating the development of robust, generalizable detection techniques. The open-source repository and comprehensive documentation are publicly available at https://github.com/sadjadeb/DetectZoo, and the package can be installed via pip install detectzoo.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript introduces DetectZoo, a first-of-its-kind extensible toolkit providing a unified API and standardized empirical pipeline for AI-generated content detection across text, audio, and image modalities. It integrates 61 detectors and 22 benchmark datasets, with each detector claimed to be self-contained, cache pretrained weights automatically, and reproduce original published results. The contribution emphasizes lowering barriers to multi-modal AI forensics through open-source code available via pip and GitHub.

Significance. If the reproduction claims hold and the wrappers preserve original detector behavior without introducing biases, the toolkit would provide a valuable engineering contribution by enabling systematic, reproducible benchmarking across modalities and reducing fragmentation from incompatible codebases. The open-source release and pip installability are positive factors for adoption.

major comments (2)
  1. [Abstract] Abstract: The central claim that 'each detector ... reproduces the original published results' is load-bearing for the utility of the unification but is asserted without any quantitative validation, comparison tables, or error analysis in the manuscript; external verification via the repository is noted but does not substitute for evidence presented in the paper.
  2. [Abstract] Abstract: The assumption that wrapping 61 existing detectors under a common API 'preserves their original behavior and performance without introducing implementation-specific biases or incompatibilities' is not tested or discussed; this directly affects whether the standardized pipeline enables fair comparisons as claimed.
minor comments (1)
  1. [Abstract] The abstract mentions 'multiple metrics through a common interface' but does not specify which metrics are standardized, which would aid clarity for readers evaluating the pipeline.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback on our manuscript. We address each major comment point by point below, proposing revisions where the concerns are valid.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The central claim that 'each detector ... reproduces the original published results' is load-bearing for the utility of the unification but is asserted without any quantitative validation, comparison tables, or error analysis in the manuscript; external verification via the repository is noted but does not substitute for evidence presented in the paper.

    Authors: We agree that the reproduction claim is central and that the manuscript would be strengthened by including quantitative validation directly in the paper rather than relying on the repository. In the revised version, we will add a new subsection (likely in Section 4 or an appendix) presenting reproduction results. This will include a table comparing key metrics (e.g., accuracy, AUC) from our implementations against the originally reported numbers for a representative subset of detectors across all three modalities, along with notes on any discrepancies and the evaluation protocol used. revision: yes

  2. Referee: [Abstract] Abstract: The assumption that wrapping 61 existing detectors under a common API 'preserves their original behavior and performance without introducing implementation-specific biases or incompatibilities' is not tested or discussed; this directly affects whether the standardized pipeline enables fair comparisons as claimed.

    Authors: The referee is correct that this assumption should be explicitly tested and discussed to support claims of fair comparisons. We will revise the manuscript to include a dedicated discussion of the wrapper design principles (e.g., preserving original preprocessing, model forward passes, and post-processing steps) and add empirical validation experiments. These will compare performance of the wrapped detectors versus their standalone original code on shared benchmark subsets, quantifying any differences and discussing mitigation strategies for potential biases. revision: yes

Circularity Check

0 steps flagged

No significant circularity; engineering integration claim with external verifiability

full rationale

The manuscript introduces a software toolkit that wraps 61 existing detectors and 22 datasets under a common API. No equations, fitted parameters, predictions, or derivations appear anywhere in the text. The central assertion—that each wrapper reproduces the original published detector behavior—is presented as an engineering requirement whose correctness is externally checkable via the linked open-source repository rather than derived internally. No self-citation chains, ansatzes, or uniqueness theorems are invoked to support the unification claim. The contribution is therefore self-contained against external benchmarks (the published detector papers and the released code).

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is a software toolkit paper with no mathematical derivations. No free parameters, axioms, or invented entities are introduced.

pith-pipeline@v0.9.1-grok · 5876 in / 1123 out tokens · 26175 ms · 2026-06-28T07:05:35.910358+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

115 extracted references · 21 canonical work pages · 8 internal anchors

  1. [1]

    Defending against neural fake news.Advances in neural information processing systems, 32, 2019

    Rowan Zellers, Ari Holtzman, Hannah Rashkin, Yonatan Bisk, Ali Farhadi, Franziska Roesner, and Yejin Choi. Defending against neural fake news.Advances in neural information processing systems, 32, 2019

    2019

  2. [2]

    Deepfake Media Generation and Detection in the Generative AI Era: A Survey and Outlook

    Florinel-Alin Croitoru, Andrei Iulian Hiji, Vlad Hondru, Nicolae-C˘at˘alin Ristea, Paul Irofti, Marius Popescu, Cristian Rusu, Radu Tudor Ionescu, Fahad Shahbaz Khan, and Mubarak Shah. Deepfake media generation and detection in the generative ai era: A survey and outlook. ArXiv, abs/2411.19537, 2024

    work page internal anchor Pith review arXiv 2024

  3. [3]

    Deep fakes: A looming challenge for privacy, democracy, and national security.Calif

    Bobby Chesney and Danielle Citron. Deep fakes: A looming challenge for privacy, democracy, and national security.Calif. L. Rev., 107:1753, 2019

    2019

  4. [4]

    Audio deepfake detection: A survey.arXiv preprint arXiv:2308.14970, 2023

    Jiangyan Yi, Chenglong Wang, Jianhua Tao, Xiaohui Zhang, Chu Yuan Zhang, and Yan Zhao. Audio deepfake detection: A survey.arXiv preprint arXiv:2308.14970, 2023

    work page arXiv 2023

  5. [5]

    Human perception of audio deepfakes

    Nicolas M Müller, Karla Pizzi, and Jennifer Williams. Human perception of audio deepfakes. InProceedings of the 1st international workshop on deepfake detection for audio multimedia, pages 85–91, 2022

    2022

  6. [6]

    Beyond a reasonable doubt? audiovisual evidence, ai manipulation, deepfakes, and the law.IEEE Transactions on Technology and Society, 5(2):156–168, 2024

    Yvonne Apolo and Katina Michael. Beyond a reasonable doubt? audiovisual evidence, ai manipulation, deepfakes, and the law.IEEE Transactions on Technology and Society, 5(2):156–168, 2024

    2024

  7. [7]

    Judicial approaches to acknowledged and unacknowl- edged ai-generated evidence.Colum

    Maura R Grossman and Paul W Grimm. Judicial approaches to acknowledged and unacknowl- edged ai-generated evidence.Colum. Sci. & Tech. L. Rev., 26:110, 2024

    2024

  8. [8]

    Detectgpt: Zero-shot machine-generated text detection using probability curvature

    Eric Mitchell, Yoonho Lee, Alexander Khazatsky, Christopher D Manning, and Chelsea Finn. Detectgpt: Zero-shot machine-generated text detection using probability curvature. In International conference on machine learning, pages 24950–24962. PMLR, 2023

    2023

  9. [9]

    Fast-detectgpt: Efficient zero-shot detection of machine-generated text via conditional probability curvature

    Guangsheng Bao, Yanbin Zhao, Zhiyang Teng, Linyi Yang, and Yue Zhang. Fast-detectgpt: Efficient zero-shot detection of machine-generated text via conditional probability curvature. arXiv preprint arXiv:2310.05130, 2023

    work page arXiv 2023

  10. [10]

    Spotting llms with binoculars: Zero-shot detection of machine-generated text.arXiv preprint arXiv:2401.12070, 2024

    Abhimanyu Hans, Avi Schwarzschild, Valeriia Cherepanova, Hamid Kazemi, Aniruddha Saha, Micah Goldblum, Jonas Geiping, and Tom Goldstein. Spotting llms with binoculars: Zero-shot detection of machine-generated text.arXiv preprint arXiv:2401.12070, 2024

    work page arXiv 2024

  11. [11]

    Towards universal fake image detectors that generalize across generative models

    Utkarsh Ojha, Yuheng Li, and Yong Jae Lee. Towards universal fake image detectors that generalize across generative models. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 24480–24489, 2023

    2023

  12. [12]

    Drct: Diffusion reconstruction contrastive training towards universal detection of diffusion generated images

    Baoying Chen, Jishen Zeng, Jianquan Yang, and Rui Yang. Drct: Diffusion reconstruction contrastive training towards universal detection of diffusion generated images. InForty-first International Conference on Machine Learning, 2024

    2024

  13. [13]

    Co-spy: Combining semantic and pixel features to detect synthetic images by ai

    Siyuan Cheng, Lingjuan Lyu, Zhenting Wang, Xiangyu Zhang, and Vikash Sehwag. Co-spy: Combining semantic and pixel features to detect synthetic images by ai. InProceedings of the Computer Vision and Pattern Recognition Conference, pages 13455–13465, 2025

    2025

  14. [14]

    End-to-end anti-spoofing with rawnet2

    Hemlata Tak, Jose Patino, Massimiliano Todisco, Andreas Nautsch, Nicholas Evans, and Anthony Larcher. End-to-end anti-spoofing with rawnet2. InICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 6369–

    2021

  15. [15]

    Aasist: Audio anti-spoofing using integrated spectro-temporal graph attention networks

    Jee-weon Jung, Hee-Soo Heo, Hemlata Tak, Hye-jin Shim, Joon Son Chung, Bong-Jin Lee, Ha-Jin Yu, and Nicholas Evans. Aasist: Audio anti-spoofing using integrated spectro-temporal graph attention networks. InICASSP 2022-2022 IEEE international conference on acoustics, speech and signal processing (ICASSP), pages 6367–6371. IEEE, 2022. 10

    2022

  16. [16]

    Climb: Class-imbalanced learning benchmark on tabular data

    Zhining Liu, Zihao Li, Ze Yang, Tianxin Wei, Jian Kang, Yada Zhu, Hendrik Hamann, Jingrui He, and Hanghang Tong. Climb: Class-imbalanced learning benchmark on tabular data. In The Thirty-ninth Annual Conference on Neural Information Processing Systems Datasets and Benchmarks Track, 2025

    2025

  17. [17]

    Robustbench: a standardized adversarial robustness benchmark

    Francesco Croce, Maksym Andriushchenko, Vikash Sehwag, Edoardo Debenedetti, Nicolas Flammarion, Mung Chiang, Prateek Mittal, and Matthias Hein. Robustbench: a standardized adversarial robustness benchmark. InThirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track, 2021

    2021

  18. [18]

    HuggingFace's Transformers: State-of-the-art Natural Language Processing

    Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pierric Cistac, Tim Rault, Rémi Louf, Morgan Funtowicz, et al. Huggingface’s trans- formers: State-of-the-art natural language processing.arXiv preprint arXiv:1910.03771, 2019

    work page internal anchor Pith review Pith/arXiv arXiv 1910

  19. [19]

    Pyod: A python toolbox for scalable outlier detection

    Yue Zhao, Zain Nasrullah, and Zheng Li. Pyod: A python toolbox for scalable outlier detection. Journal of Machine Learning Research, 20(96):1–7, 2019

    2019

  20. [20]

    Release Strategies and the Social Impacts of Language Models

    Irene Solaiman, Miles Brundage, Jack Clark, Amanda Askell, Ariel Herbert-V oss, Jeff Wu, Alec Radford, Gretchen Krueger, Jong Wook Kim, Sarah Kreps, et al. Release strategies and the social impacts of language models.arXiv preprint arXiv:1908.09203, 2019

    work page internal anchor Pith review Pith/arXiv arXiv 1908

  21. [21]

    Gltr: Statistical detection and visualization of generated text

    Sebastian Gehrmann, Hendrik Strobelt, and Alexander M Rush. Gltr: Statistical detection and visualization of generated text. InProceedings of the 57th annual meeting of the association for computational linguistics: system demonstrations, pages 111–116, 2019

    2019

  22. [22]

    Detecting fake content with relative entropy scoring.Pan, 8(27-31):4, 2008

    Thomas Lavergne, Tanguy Urvoy, and François Yvon. Detecting fake content with relative entropy scoring.Pan, 8(27-31):4, 2008

    2008

  23. [23]

    Detectllm: Leveraging log rank information for zero-shot detection of machine-generated text

    Jinyan Su, Terry Zhuo, Di Wang, and Preslav Nakov. Detectllm: Leveraging log rank information for zero-shot detection of machine-generated text. InFindings of the Association for Computational Linguistics: EMNLP 2023, pages 12395–12412, 2023

    2023

  24. [24]

    Dna-gpt: Divergent n-gram analysis for training-free detection of gpt-generated text.arXiv preprint arXiv:2305.17359, 2023

    Xianjun Yang, Wei Cheng, Yue Wu, Linda Petzold, William Yang Wang, and Haifeng Chen. Dna-gpt: Divergent n-gram analysis for training-free detection of gpt-generated text.arXiv preprint arXiv:2305.17359, 2023

    work page arXiv 2023

  25. [25]

    Intrinsic dimen- sion estimation for robust detection of ai-generated texts.Advances in Neural Information Processing Systems, 36:39257–39276, 2023

    Eduard Tulchinskii, Kristian Kuznetsov, Laida Kushnareva, Daniil Cherniavskii, Sergey Nikolenko, Evgeny Burnaev, Serguei Barannikov, and Irina Piontkovskaya. Intrinsic dimen- sion estimation for robust detection of ai-generated texts.Advances in Neural Information Processing Systems, 36:39257–39276, 2023

    2023

  26. [26]

    Radar: Robust ai-text detection via adversarial learning.Advances in neural information processing systems, 36:15077–15095, 2023

    Xiaomeng Hu, Pin-Yu Chen, and Tsung-Yi Ho. Radar: Robust ai-text detection via adversarial learning.Advances in neural information processing systems, 36:15077–15095, 2023

    2023

  27. [27]

    Remodetect: Reward models recognize aligned llm’s generations.Advances in Neural Information Processing Systems, 37:2886–2913, 2024

    Hyunseok Lee, Jihoon Tack, and Jinwoo Shin. Remodetect: Reward models recognize aligned llm’s generations.Advances in Neural Information Processing Systems, 37:2886–2913, 2024

    2024

  28. [28]

    Cnn- generated images are surprisingly easy to spot

    Sheng-Yu Wang, Oliver Wang, Richard Zhang, Andrew Owens, and Alexei A Efros. Cnn- generated images are surprisingly easy to spot... for now. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 8695–8704, 2020

    2020

  29. [29]

    Rethinking the up-sampling operations in cnn-based generative network for generalizable deepfake detection

    Chuangchuang Tan, Yao Zhao, Shikui Wei, Guanghua Gu, Ping Liu, and Yunchao Wei. Rethinking the up-sampling operations in cnn-based generative network for generalizable deepfake detection. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 28130–28139, 2024

    2024

  30. [30]

    Frequency-aware deepfake detection: Improving generalizability through frequency space domain learning

    Chuangchuang Tan, Yao Zhao, Shikui Wei, Guanghua Gu, Ping Liu, and Yunchao Wei. Frequency-aware deepfake detection: Improving generalizability through frequency space domain learning. InProceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 5052–5060, 2024. 11

    2024

  31. [31]

    Improving synthetic image detection towards generalization: An image transformation perspective

    Ouxiang Li, Jiayin Cai, Yanbin Hao, Xiaolong Jiang, Yao Hu, and Fuli Feng. Improving synthetic image detection towards generalization: An image transformation perspective. In Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V . 1, pages 2405–2414, 2025

    2025

  32. [32]

    Learning on gradients: Generalized artifacts representation for gan-generated images detection

    Chuangchuang Tan, Yao Zhao, Shikui Wei, Guanghua Gu, and Yunchao Wei. Learning on gradients: Generalized artifacts representation for gan-generated images detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 12105–12114, 2023

    2023

  33. [33]

    Aeroblade: Training-free detection of latent diffusion images using autoencoder reconstruction error

    Jonas Ricker, Denis Lukovnikov, and Asja Fischer. Aeroblade: Training-free detection of latent diffusion images using autoencoder reconstruction error. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 9130–9140, 2024

    2024

  34. [34]

    The unrea- sonable effectiveness of deep features as a perceptual metric

    Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shechtman, and Oliver Wang. The unrea- sonable effectiveness of deep features as a perceptual metric. InProceedings of the IEEE conference on computer vision and pattern recognition, pages 586–595, 2018

    2018

  35. [35]

    Forgery-aware adaptive transformer for generalizable synthetic image detection

    Huan Liu, Zichang Tan, Chuangchuang Tan, Yunchao Wei, Jingdong Wang, and Yao Zhao. Forgery-aware adaptive transformer for generalizable synthetic image detection. InPro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 10770–10780, 2024

    2024

  36. [36]

    C2p-clip: Injecting category common prompt in clip to enhance generalization in deepfake detection

    Chuangchuang Tan, Renshuai Tao, Huan Liu, Guanghua Gu, Baoyuan Wu, Yao Zhao, and Yunchao Wei. C2p-clip: Injecting category common prompt in clip to enhance generalization in deepfake detection. InProceedings of the AAAI Conference on Artificial Intelligence, volume 39, pages 7184–7192, 2025

    2025

  37. [37]

    A sanity check for ai-generated image detection

    Shilin Yan, Ouxiang Li, Jiayin Cai, Yanbin Hao, Xiaolong Jiang, Yao Hu, and Weidi Xie. A sanity check for ai-generated image detection. InThe Thirteenth International Conference on Learning Representations

  38. [38]

    Improved DeepFake Detection Using Whisper Features

    Piotr Kawa, Marcin Plata, Michał Czuba, Piotr Szyma´nski, and Piotr Syga. Improved DeepFake Detection Using Whisper Features. InProc. INTERSPEECH 2023, pages 4009–4013, 2023

    2023

  39. [39]

    RAID: A shared benchmark for robust evaluation of machine-generated text detectors

    Liam Dugan, Alyssa Hwang, Filip Trhlík, Andrew Zhu, Josh Magnus Ludan, Hainiu Xu, Daphne Ippolito, and Chris Callison-Burch. RAID: A shared benchmark for robust evaluation of machine-generated text detectors. InProceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 12463–12492, Bangkok, Thail...

    2024

  40. [40]

    Turingbench: A bench- mark environment for turing test in the age of neural text generation

    Adaku Uchendu, Zeyu Ma, Thai Le, Rui Zhang, and Dongwon Lee. Turingbench: A bench- mark environment for turing test in the age of neural text generation. InFindings of the association for computational linguistics: EMNLP 2021, pages 2001–2016, 2021

    2021

  41. [41]

    M4: Multi-generator, multi- domain, and multi-lingual black-box machine-generated text detection

    Yuxia Wang, Jonibek Mansurov, Petar Ivanov, Jinyan Su, Artem Shelmanov, Akim Tsvigun, Chenxi Whitehouse, Osama Mohammed Afzal, Tarek Mahmoud, Toru Sasaki, Thomas Arnold, Alham Aji, Nizar Habash, Iryna Gurevych, and Preslav Nakov. M4: Multi-generator, multi- domain, and multi-lingual black-box machine-generated text detection. InProceedings of the 18th Con...

    2024

  42. [42]

    MGTBench: Benchmarking Machine-Generated Text Detection

    Xinlei He, Xinyue Shen, Zeyuan Chen, Michael Backes, and Yang Zhang. MGTBench: Benchmarking Machine-Generated Text Detection. InACM SIGSAC Conference on Computer and Communications Security (CCS). ACM, 2024

    2024

  43. [43]

    Deepfakebench: A comprehensive benchmark of deepfake detection.Advances in Neural Information Processing Systems, 36:4534–4565, 2023

    Zhiyuan Yan, Yong Zhang, Xinhang Yuan, Siwei Lyu, and Baoyuan Wu. Deepfakebench: A comprehensive benchmark of deepfake detection.Advances in Neural Information Processing Systems, 36:4534–4565, 2023. 12

    2023

  44. [44]

    Genimage: A million-scale benchmark for detecting ai-generated image.Advances in neural information processing systems, 36:77771–77782, 2023

    Mingjian Zhu, Hanting Chen, Qiangyu Yan, Xudong Huang, Guanyu Lin, Wei Li, Zhijun Tu, Hailin Hu, Jie Hu, and Yunhe Wang. Genimage: A million-scale benchmark for detecting ai-generated image.Advances in neural information processing systems, 36:77771–77782, 2023

    2023

  45. [45]

    Junhao Zhuang, Yanhong Zeng, Wenran Liu, Chun Yuan, and Kai Chen

    Nan Zhong, Yiran Xu, Sheng Li, Zhenxing Qian, and Xinpeng Zhang. Patchcraft: Exploring texture patch for efficient ai-generated image detection.arXiv preprint arXiv:2311.12397, 2023

    work page arXiv 2023

  46. [46]

    ASVspoof 2019: Future Horizons in Spoofed and Fake Audio Detection

    Massimiliano Todisco, Xin Wang, Ville Vestman, Md Sahidullah, Héctor Delgado, Andreas Nautsch, Junichi Yamagishi, Nicholas Evans, Tomi Kinnunen, and Kong Aik Lee. Asvspoof 2019: Future horizons in spoofed and fake audio detection.arXiv preprint arXiv:1904.05441, 2019

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  47. [47]

    ASVspoof 2021: accelerating progress in spoofed and deepfake speech detection

    Junichi Yamagishi, Xin Wang, Massimiliano Todisco, Md Sahidullah, Jose Patino, Andreas Nautsch, Xuechen Liu, Kong Aik Lee, Tomi Kinnunen, Nicholas Evans, and Héctor Delgado. ASVspoof 2021: accelerating progress in spoofed and deepfake speech detection. InProc. ASVspoof Challenge workshop, pages 47–54, 2021

    2021

  48. [48]

    Mage: Machine-generated text detection in the wild

    Yafu Li, Qintong Li, Leyang Cui, Wei Bi, Zhilin Wang, Longyue Wang, Linyi Yang, Shuming Shi, and Yue Zhang. Mage: Machine-generated text detection in the wild. InProceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 36–53, 2024

    2024

  49. [49]

    Detectrl: Benchmarking llm-generated text detection in real-world scenarios.arXiv preprint arXiv:2410.23746, 2024

    Junchao Wu, Runzhe Zhan, Derek F Wong, Shu Yang, Xinyi Yang, Yulin Yuan, and Lidia S Chao. Detectrl: Benchmarking llm-generated text detection in real-world scenarios.arXiv preprint arXiv:2410.23746, 2024

    work page arXiv 2024

  50. [50]

    Is artificial intelligence generated image detection a solved problem? InThe Thirty-ninth Annual Conference on Neural Information Processing Systems Datasets and Benchmarks Track

    Ziqiang Li, Jiazhen Yan, Ziwen He, Kai Zeng, Weiwei Jiang, Lizhi Xiong, and Zhangjie Fu. Is artificial intelligence generated image detection a solved problem? InThe Thirty-ninth Annual Conference on Neural Information Processing Systems Datasets and Benchmarks Track

  51. [51]

    Training-free llm-generated text detection by mining token probability sequences.arXiv preprint arXiv:2410.06072, 2024

    Yihuai Xu, Yongwei Wang, Yifei Bi, Huangsen Cao, Zhouhan Lin, Yu Zhao, and Fei Wu. Training-free llm-generated text detection by mining token probability sequences.arXiv preprint arXiv:2410.06072, 2024

    work page arXiv 2024

  52. [52]

    Who wrote this? the key to zero-shot llm-generated text detection is gecscore

    Junchao Wu, Runzhe Zhan, Derek F Wong, Shu Yang, Xuebo Liu, Lidia S Chao, and Min Zhang. Who wrote this? the key to zero-shot llm-generated text detection is gecscore. In Proceedings of the 31st International Conference on Computational Linguistics, pages 10275– 10292, 2025

    2025

  53. [53]

    Biscope: Ai-generated text detection by checking memorization of preceding tokens.Advances in Neural Information Processing Systems, 37:104065–104090, 2024

    Hanxi Guo, Siyuan Cheng, Xiaolong Jin, Zhuo Zhang, Kaiyuan Zhang, Guanhong Tao, Guangyu Shen, and Xiangyu Zhang. Biscope: Ai-generated text detection by checking memorization of preceding tokens.Advances in Neural Information Processing Systems, 37:104065–104090, 2024

    2024

  54. [54]

    Adadetectgpt: Adaptive detection of llm-generated text with statistical guarantees

    Hongyi Zhou, Jin Zhu, Pingfan Su, Kai Ye, Ying Yang, Shakeel A O B Gavioli-Akilagun, and Chengchun Shi. Adadetectgpt: Adaptive detection of llm-generated text with statistical guarantees. InThe Thirty-Ninth Annual Conference on Neural Information Processing Systems, 2025

    2025

  55. [55]

    Glimpse: Enabling white-box methods to use proprietary models for zero-shot llm-generated text detection

    Guangsheng Bao, Yanbin Zhao, Juncai He, and Yue Zhang. Glimpse: Enabling white-box methods to use proprietary models for zero-shot llm-generated text detection. 2025

    2025

  56. [56]

    Dna- detectllm: Unveiling ai-generated text via a dna-inspired mutation-repair paradigm

    Xiaowei Zhu, Yubing Ren, Fang Fang, Qingfeng Tan, Shi Wang, and Yanan Cao. Dna- detectllm: Unveiling ai-generated text via a dna-inspired mutation-repair paradigm. InThe Thirty-ninth Annual Conference on Neural Information Processing Systems, 2025

    2025

  57. [57]

    Beat LLMs at their own game: Zero-shot LLM- generated text detection via querying ChatGPT

    Biru Zhu, Lifan Yuan, Ganqu Cui, Yangyi Chen, Chong Fu, Bingxiang He, Yangdong Deng, Zhiyuan Liu, Maosong Sun, and Ming Gu. Beat LLMs at their own game: Zero-shot LLM- generated text detection via querying ChatGPT. InProceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 7470–7483, Singapore, December

    2023

  58. [58]

    Association for Computational Linguistics. 13

  59. [59]

    Raidar: generative AI detection via rewriting

    Chengzhi Mao, Carl V ondrick, Hao Wang, and Junfeng Yang. Raidar: generative AI detection via rewriting. InThe Twelfth International Conference on Learning Representations, 2024

    2024

  60. [60]

    Ghostbuster: Detecting text ghostwritten by large language models

    Vivek Verma, Eve Fleisig, Nicholas Tomlin, and Dan Klein. Ghostbuster: Detecting text ghostwritten by large language models. InProceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Tech- nologies (Volume 1: Long Papers), pages 1702–1717, 2024

    2024

  61. [61]

    Zero-shot detection of LLM-generated text using token cohe- siveness

    Shixuan Ma and Quan Wang. Zero-shot detection of LLM-generated text using token cohe- siveness. In Yaser Al-Onaizan, Mohit Bansal, and Yun-Nung Chen, editors,Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 17538–17553, Miami, Florida, USA, November 2024. Association for Computational Linguistics

    2024

  62. [62]

    Ipad: Inverse prompt for ai detection-a robust and interpretable llm-generated text detector.arXiv preprint arXiv:2502.15902, 2025

    Zheng Chen, Yushi Feng, Jisheng Dang, Yue Deng, Changyang He, Hongxi Pu, Haoxuan Li, and Bo Li. Ipad: Inverse prompt for ai detection-a robust and interpretable llm-generated text detector.arXiv preprint arXiv:2502.15902, 2025

    work page arXiv 2025

  63. [63]

    Text fluoroscopy: Detecting llm-generated text through intrinsic features

    Xiao Yu, Kejiang Chen, Qi Yang, Weiming Zhang, and Nenghai Yu. Text fluoroscopy: Detecting llm-generated text through intrinsic features. InProceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 15838–15846, 2024

    2024

  64. [64]

    Coco: Coherence-enhanced machine-generated text detection under low resource with contrastive learning

    Xiaoming Liu, Zhaohan Zhang, Yichen Wang, Hang Pu, Yu Lan, and Chao Shen. Coco: Coherence-enhanced machine-generated text detection under low resource with contrastive learning. Inproceedings of the 2023 conference on empirical methods in natural language processing, pages 16167–16188, 2023

    2023

  65. [65]

    Imitate before detect: Aligning machine stylistic preference for machine-revised text detection

    Jiaqi Chen, Xiaoye Zhu, Tianyang Liu, Ying Chen, Chen Xinhui, Yiwen Yuan, Chak Tou Leong, Zuchao Li, Long Tang, Lei Zhang, et al. Imitate before detect: Aligning machine stylistic preference for machine-revised text detection. InProceedings of the AAAI Conference on Artificial Intelligence, volume 39, pages 23559–23567, 2025

    2025

  66. [66]

    Detective: Detecting ai-generated text via multi-level contrastive learn- ing.Advances in Neural Information Processing Systems, 37:88320–88347, 2024

    Xun Guo, Shan Zhang, Yongxin He, Ting Zhang, Wanquan Feng, Haibin Huang, and Chongyang Ma. Detective: Detecting ai-generated text via multi-level contrastive learn- ing.Advances in Neural Information Processing Systems, 37:88320–88347, 2024

    2024

  67. [67]

    Zero-Shot Detection of LLM-Generated Text via Implicit Reward Model

    Runheng Liu, Heyan Huang, Xingchen Xiao, and Zhijing Wu. Zero-shot detection of llm- generated text via implicit reward model.arXiv preprint arXiv:2604.21223, 2026

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  68. [68]

    Human texts are outliers: Detecting llm-generated texts via out-of-distribution detection.arXiv preprint arXiv:2510.08602, 2025

    Cong Zeng, Shengkun Tang, Yuanzhou Chen, Zhiqiang Shen, Wenchao Yu, Xujiang Zhao, Haifeng Chen, Wei Cheng, and Zhiqiang Xu. Human texts are outliers: Detecting llm-generated texts via out-of-distribution detection.arXiv preprint arXiv:2510.08602, 2025

    work page arXiv 2025

  69. [69]

    Progressive growing of gans for improved quality, stability, and variation

    Tero Karras, Timo Aila, Samuli Laine, and Jaakko Lehtinen. Progressive growing of gans for improved quality, stability, and variation. InInternational Conference on Learning Represen- tations, 2018

    2018

  70. [70]

    Dˆ 3: scaling up deepfake detection by learning from discrepancy

    Yongqi Yang, Zhihao Qian, Ye Zhu, Olga Russakovsky, and Yu Wu. Dˆ 3: scaling up deepfake detection by learning from discrepancy. InProceedings of the Computer Vision and Pattern Recognition Conference, pages 23850–23859, 2025

    2025

  71. [71]

    Manifold induced biases for zero-shot and few-shot detection of generated images

    Jonathan Brokman, Amit Giloni, Omer Hofman, Roman Vainshtein, Hisashi Kojima, and Guy Gilboa. Manifold induced biases for zero-shot and few-shot detection of generated images. In 13th International Conference on Learning Representations, ICLR 2025, pages 12803–12828. International Conference on Learning Representations, ICLR, 2025

    2025

  72. [72]

    End- to-end spectro-temporal graph attention networks for speaker verification anti-spoofing and speech deepfake detection

    Hemlata Tak, Jee-weon Jung, Jose Patino, Massimiliano Todisco, and Nicholas Evans. End- to-end spectro-temporal graph attention networks for speaker verification anti-spoofing and speech deepfake detection. InProc. ASVspoof Challenge workshop, pages 1–8, 2021

    2021

  73. [73]

    Towards end-to-end synthetic speech detection.IEEE Signal Processing Letters, 28:1265–1269, 2021

    Guang Hua, Andrew Beng Jin Teoh, and Haijian Zhang. Towards end-to-end synthetic speech detection.IEEE Signal Processing Letters, 28:1265–1269, 2021. 14

    2021

  74. [74]

    SAMO: Speaker attractor multi-center one-class learning for voice anti-spoofing

    Siying Ding, You Zhang, and Zhiyao Duan. SAMO: Speaker attractor multi-center one-class learning for voice anti-spoofing. InICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1–5. IEEE, 2023

    2023

  75. [75]

    AST: Audio spectrogram transformer

    Yuan Gong, Yu-An Chung, and James Glass. AST: Audio spectrogram transformer. InProc. Interspeech 2021, pages 571–575, 2021

    2021

  76. [76]

    Post-training for deepfake speech detection.arXiv preprint arXiv:2506.21090, 2025

    Wanying Ge, Xin Wang, Junichi Yamagishi, and Nicholas Evans. Post-training for deepfake speech detection.arXiv preprint arXiv:2506.21090, 2025

    work page arXiv 2025

  77. [77]

    Audio deepfake detection with self-supervised XLS-R and sensitive layer selection

    Qishan Zhang, Jian Ye, Yue Lu, and Huan Liu. Audio deepfake detection with self-supervised XLS-R and sensitive layer selection. InProceedings of the 32nd ACM International Conference on Multimedia, pages 1–10. ACM, 2024

    2024

  78. [78]

    How close is chatgpt to human experts? comparison corpus, evaluation, and detection.arXiv preprint arxiv:2301.07597, 2023

    Biyang Guo, Xin Zhang, Ziyuan Wang, Minqi Jiang, Jinran Nie, Yuxuan Ding, Jianwei Yue, and Yupeng Wu. How close is chatgpt to human experts? comparison corpus, evaluation, and detection.arXiv preprint arxiv:2301.07597, 2023

    work page arXiv 2023

  79. [79]

    Hc3 plus: A semantic- invariant human chatgpt comparison corpus.arXiv preprint arXiv:2309.02731, 2023

    Zhenpeng Su, Xing Wu, Wei Zhou, Guangyuan Ma, and Songlin Hu. Hc3 plus: A semantic- invariant human chatgpt comparison corpus.arXiv preprint arXiv:2309.02731, 2023

    work page arXiv 2023

  80. [80]

    Cheat: A large-scale dataset for detecting chatgpt-written abstracts.IEEE Transactions on Big Data, 11(3):898–906, 2025

    Peipeng Yu, Jiahan Chen, Xuan Feng, and Zhihua Xia. Cheat: A large-scale dataset for detecting chatgpt-written abstracts.IEEE Transactions on Big Data, 11(3):898–906, 2025

    2025

Showing first 80 references.