Robustness of Machine-Generated Text Detection

Overview

Robustness of machine-generated text detection asks whether systems can distinguish AI-written from human-written text beyond familiar benchmark conditions, which matters for fair screening and trustworthy provenance claims. Methods range from supervised transformers and linguistic or stylometric classifiers to zero-shot likelihood and entropy scores, representation probes, and embedded watermarks. Evaluations test domain and generator shifts, multilingual and low-resource writing, paraphrase and style attacks, mixed human–AI edits, and low false-positive rates on benchmarks such as RAID, MAGE, and DetectRL. Strong in-domain scores often fail to transfer, while editing, formatting artifacts, and rewriting can both evade detectors and trigger false accusations; the central tension is improving coverage without treating text-only scores as proof of authorship.

Latest papers

CardsList

104 papers16 papers in the last four weeks, down 30% on the four weeks before. 0.1% of all new papers.

  1. Who Can You Hide Behind? Author-Conditioned Style Transfer and the Evasion of a Commercial AI-Text Detector

    Nov 15, 2026Mohd Ruhul Ameen, M. Saifuzzaman Rafat, Akif IslamRobustness of Machine-Generated Text DetectionAI-Generated Text Detection

  2. Lasso: Towards Sustainable AI-Generated Media Detection with Intel QAT On-Chip Compression

    Nov 13, 2026Darsh Asher, Joshua Kalyanapu, Kaushal Mhapsekar +4AI-Generated Text DetectionDeepfake Detection

  3. SPADE: Soft Prompts for Ai-text DEtection

    Nov 12, 2026Alex Jiang, Didier Schwab, Chuyuan LiAI-Generated Text DetectionRobustness of Machine-Generated Text Detection

  4. FIBER: Bayesian Fusion of Fast-DetectGPT and IRT-Informed Detector Evidence for LLM-Generated Text Detection

    Nov 11, 2026Yan Li, Guangsheng Bao, Zhipeng Chen +2AI-Generated Text DetectionRobustness of Machine-Generated Text Detection

  5. Detecting AI-Authored Text in Authentic Student Writing: Evidence from PISA 2025

    Nov 9, 2026Nils-Jonathan Schaller, Thorben Jansen, Olaf Köller +1AI-Generated Text DetectionRobustness of Machine-Generated Text Detection

  6. Specificity Bias and the Wikipedia Paradox: An xAI Analysis of AI Text Detectors under Domain Shift

    Nov 2, 2026Nico Morloher, Johannes Reschke, Mohamed MadyAI-Generated Text DetectionRobustness of Machine-Generated Text Detection

  7. Towards Fine-grained and Robust Detection of Human-AI Co-authorship

    Nov 1, 2026Ziruo Zhao, Jie Fu, Wendy Hui Wang +1Robustness of Machine-Generated Text Detection

  8. Humanized but Still True? Auditing Facts and Detection in AI-Text Humanizers

    Oct 30, 2026Ziyuan Wang, Genliang Zhu, Xiaobo ChenLanguage Model AuditingText Generation

  9. NormDetect: Anchored Calibration for Low-False-Positive Detection of Machine-Generated Web Text

    Oct 30, 2026Hanshuai Cui, Fanshuai Meng, Zhiqing TangAI-Generated Text DetectionRobustness of Machine-Generated Text Detection

  10. TRIDENT and SHAKTI: A Framework and Benchmark for Hindi AI-Text Detection

    Oct 30, 2026Shifali Agrahari, Sujit Kumar, Sanasam Ranbir SinghAI-Generated Text DetectionRobustness of Machine-Generated Text Detection

  11. Robust LLM-Generated Text Detection through Matched Conditional-Response Probing

    Oct 28, 2026Junfeng Hao, Wenxian Wang, Tao Leng +3AI-Generated Text DetectionRobustness of Machine-Generated Text Detection

  12. Building Auditable AI Text Detectors with Sparse Autoencoders

    Oct 21, 2026Liam Dugan, Callum McDougall, Matthew Denton +3AI-Generated Text DetectionExplainable Text Classification

  13. A Unified Attack Framework against Statistical Generative Watermarks for Large Language Models

    Oct 20, 2026Romina Omidi, Nima Naderloui, Yuan Hong +1LLM WatermarkingRobustness of Machine-Generated Text Detection

  14. Beyond Surface Cues: Robust LLM-Generated Korean Text Detection

    Oct 16, 2026SungHo Kim, SangKeun LeeAI-Generated Text DetectionRobustness of Machine-Generated Text Detection

  15. Multimodal AI Detection for Math-Heavy Scientific Text

    Oct 15, 2026Yixuan Huang, Danish Pruthi, Bhuwan DhingraAI-Generated Text DetectionRobustness of Machine-Generated Text Detection

  16. Unchecked or Falsely Accused? Evaluating AI-Generated Text Detectors on Kazakh

    Oct 11, 2026Mansur OzamanAI-Generated Text DetectionRobustness of Machine-Generated Text Detection

  17. Measuring Artifact Reliance in Machine-Generated Text Detectors

    Oct 9, 2026Anastasia Voznyuk, Andrey GrabovoySpurious CorrelationAI-Generated Text Detection

  18. EASE: Entropy-Adaptive Distribution Shaping for Evading AI-generated Text Detectors

    Oct 7, 2026Jicheng Zhou, Kahim Wong, Jialong Wang +1Adversarial Text AttacksRobustness of Machine-Generated Text Detection

  19. Constitution-Guided Watermarking

    Oct 7, 2026Toluwani Aremu, Samuele Poppi, Nils LukasLLM WatermarkingRobust Watermarking

  20. Watermarking: from Impossibility to Auditable Compliance

    Oct 5, 2026Fernando Delbianco, Fernando Tohmé, Hugo AcciarriText WatermarkingRobust Watermarking