mlsanityData to 7 Oct 2026

Automatic Detection of Generated Text is Easiest when Humans are Fooled

arXiv 1911.00650 ↗First posted 2 Nov 2019cs.CLcited by 260

Not on the map. Why?

Abstract (arXiv)

Recent advancements in neural language modelling make it possible to rapidly generate vast amounts of human-sounding text. The capabilities of humans and automatic discriminators to detect machine-generated text have been a large source of research interest, but humans and machines rely on different cues to make their decisions. Here, we perform careful benchmarking and analysis of three popular sampling-based decoding strategies---top-k, nucleus sampling, and untruncated random sampling---and show that improvements in decoding methods have primarily optimized for fooling humans. This comes at the expense of introducing statistical abnormalities that make detection easy for automatic systems. We also show that though both human and automatic detector performance improve with longer excerpt length, even multi-sentence excerpts can fool expert human raters over 30% of the time. Our findings reveal the importance of using both human and automatic detectors to assess the humanness of text generation systems.

Key facts. Cited by 260 papers in the map (data through 7 Oct 2026). First posted on arXiv 2 Nov 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 13; 2022: 23; 2023: 64; 2024: 59; 2025: 56; 2026: 45.

260 papers.

Sort:

By area:

  1. 2023DetectGPT: Zero-Shot Machine-Generated Text Detection using Probability Curvaturecited by 705
  2. 2021MAUVE: Measuring the Gap Between Neural Text and Human Text using Divergence Frontierscited by 466
  3. 2023Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defensecited by 401
  4. 2023Can AI-Generated Text be Reliably Detected?cited by 372
  5. 2021All That's 'Human' Is Not Gold: Evaluating Human Evaluation of Generated Textcited by 287
  6. 2023Fast-DetectGPT: Efficient Zero-Shot Detection of Machine-Generated Text via Conditional Probability Curvaturecited by 278
  7. 2023DetectLLM: Leveraging Log Rank Information for Zero-Shot Detection of Machine-Generated Textcited by 207
  8. 2023RADAR: Robust AI-Text Detection via Adversarial Learningcited by 187
  9. 2023GPT detectors are biased against non-native English writerscited by 175
  10. 2024Monitoring AI-Modified Content at Scale: A Case Study on the Impact of ChatGPT on AI Conference Peer Reviewscited by 160
  11. 2024RAID: A Shared Benchmark for Robust Evaluation of Machine-Generated Text Detectorscited by 157
  12. 2022Repairing the Cracked Foundation: A Survey of Obstacles in Evaluation Practices for Generated Textcited by 154
38 more
  1. 2023DNA-GPT: Divergent N-Gram Analysis for Training-Free Detection of GPT-Generated Textcited by 127
  2. 2023The Science of Detecting LLM-Generated Textscited by 126
  3. 2022Truncation Sampling as Language Model Desmoothingcited by 121
  4. 2021TURINGBENCH: A Benchmark Environment for Turing Test in the Age of Neural Text Generationcited by 118
  5. 2023M4: Multi-generator, Multi-domain, and Multi-lingual Black-Box Machine-Generated Text Detectioncited by 118
  6. 2023Who Wrote this Code? Watermarking for Code Generationcited by 116
  7. 2024Mapping the Increasing Use of LLMs in Scientific Paperscited by 112
  8. 2022Human heuristics for AI-generated language are flawedcited by 104
  9. 2023A Survey on LLM-Generated Text Detection: Necessity, Methods, and Future Directionscited by 101
  10. 2021The Perils of Using Mechanical Turk to Evaluate Open-Ended Text Generationcited by 96
  11. 2023Language Model Behavior: A Comprehensive Surveycited by 92
  12. 2023MGTBench: Benchmarking Machine-Generated Text Detectioncited by 90
  13. 2023Ghostbuster: Detecting Text Ghostwritten by Large Language Modelscited by 90
  14. 2023Identifying and Mitigating the Security Risks of Generative AIcited by 90
  15. 2023Testing of Detection Tools for AI-Generated Textcited by 86
  16. 2023A Survey of GPT-3 Family Large Language Models Including ChatGPT and GPT-4cited by 81
  17. 2022Language Generation Models Can Cause Harm: So What Can We Do About It? An Actionable Surveycited by 80
  18. 2023OUTFOX: LLM-Generated Essay Detection Through In-Context Learning with Adversarially Generated Examplescited by 80
  19. 2023Three Bricks to Consolidate Watermarks for Large Language Modelscited by 80
  20. 2022Machine Generated Text: A Comprehensive Survey of Threat Models and Detection Methodscited by 75
  21. 2022Do Language Models Plagiarize?cited by 72
  22. 2023A Survey on Detection of LLMs-Generated Contentcited by 70
  23. 2022Real or Fake Text?: Investigating Human Ability to Detect Boundaries Between Human-Written and Machine-Generated Textcited by 63
  24. 2023Contrasting Linguistic Patterns in Human and LLM-Generated News Textcited by 62
  25. 2023Mark My Words: Analyzing and Evaluating Language Model Watermarkscited by 62
  26. 2023Stylometric Detection of AI-Generated Text in Twitter Timelinescited by 57
  27. 2023SeqXGPT: Sentence-Level AI-Generated Text Detectioncited by 55
  28. 2024Raidar: geneRative AI Detection viA Rewritingcited by 51
  29. 2021Artificial Text Detection via Examining the Topology of Attention Mapscited by 49
  30. 2022Synthetic Disinformation Attacks on Automated Fact Verification Systemscited by 44
  31. 2024M4GT-Bench: Evaluation Benchmark for Black-Box Machine-Generated Text Detectioncited by 43
  32. 2022Attribution and Obfuscation of Neural Text Authorship: A Data Mining Perspectivecited by 40
  33. 2023GPT-Sentinel: Distinguishing Human and ChatGPT Generated Contentcited by 40
  34. 2024Few-Shot Detection of Machine-Generated Text using Style Representationscited by 40
  35. 2024AI as Humanity's Salieri: Quantifying Linguistic Creativity of Language Models via Systematic Attribution of Machine Text against Web Textcited by 40
  36. 2023CHEAT: A Large-scale Dataset for Detecting ChatGPT-writtEn AbsTractscited by 39
  37. 2023Fighting Fire with Fire: Can ChatGPT Detect AI-generated Text?cited by 39
  38. 2025People who frequently use ChatGPT for writing tasks are accurate and robust detectors of AI-generated textcited by 39

50 most cited of 260 All 260 (JSON)