Trolling-oriented generations
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Harmful content detection | Trolling-oriented generations DeepSeek-Llama 70B | Accuracy16.24 | 4 | |
| Harmful content detection | Trolling-oriented generations Llama-3.1 70B | Accuracy26.04 | 4 | |
| Harmful content detection | Trolling-oriented generations GPT-4o | Accuracy19.88 | 4 |
Showing 3 of 3 rows