Binary Classification (Human vs Assistive) on Wikipedia
96AUCRole-Aware AI
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Role-Aware AIGenerator Model=GPT-2, Generator Parameters=124 million2026.05 | 96 | 91 | |
| Role-Aware AIGenerator Model=LLaMA-3-Instruct, Generator Parameters=3 billion2026.05 | 96 | 90 | |
| RoBERTaGenerator Model=GPT-2, Generator Parameters=124 million2026.05 | 94 | 88 | |
| LogRankGenerator Model=GPT-2, Generator Parameters=124 million2026.05 | 84 | 77 | |
| LogLikelihoodGenerator Model=LLaMA-3-Instruct, Generator Parameters=3 billion2026.05 | 82 | 74 | |
| LogRankGenerator Model=LLaMA-3-Instruct, Generator Parameters=3 billion2026.05 | 80 | 72 | |
| EntropyGenerator Model=LLaMA-3-Instruct, Generator Parameters=3 billion2026.05 | 77 | 69 | |
| RoBERTaGenerator Model=LLaMA-3-Instruct, Generator Parameters=3 billion2026.05 | 72 | 70 | |
| CurvatureGenerator Model=LLaMA-3-Instruct, Generator Parameters=3 billion2026.05 | 71 | 65 | |
| CurvatureGenerator Model=GPT-2, Generator Parameters=124 million2026.05 | 62 | 59 | |
| EntropyGenerator Model=GPT-2, Generator Parameters=124 million2026.05 | 55 | 52 | |
| LogLikelihoodGenerator Model=GPT-2, Generator Parameters=124 million2026.05 | 24 | 50 |