| Category | How Detectors Work | Evasion Techniques |
|---|---|---|
| Patterns & statistical models | NLP, perplexity, burstiness, embeddings, classification | Rewrite, changed style, add personality |
| Watermarking | Hidden token biases when making something | Change words, use changes |
| Database comparison | Find text that was made before by retrieval | Make new words |
| Adversarial training | Reinforcement learning to make things better against detectors | Use systems like AuthorMist |
| Stylometric hiding | Hide style by translation or copying | Adversarial stylometry, changing-word models |
Last active
August 19, 2025 10:47
-
-
Save alehano/aa0aa924937fac1cec883af8d1c22a51 to your computer and use it in GitHub Desktop.
AI detection methods
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment