| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| LLaMA 8B 3.1 | ArtPrompt | Mean Perplexity1.99 | 10 | 16d ago | |
| LLaMA 2 7B | Mean Perplexity (PPL)2.75 | 10 | 16d ago | ||
| LLaMA 7B | Mean PPL2.78 | 10 | 16d ago | ||
| WildGuard 7B | Mean Perplexity2.33 | 10 | 16d ago | ||
| LLaMA Guard 8B 3.1 | ArtPrompt | Mean PPL2.16 | 10 | 16d ago | |
| LLaMA Guard 2 8B | ArtPrompt | PPL Mean2.34 | 10 | 16d ago | |
| LLaMA Guard 7B | Mean Perplexity (PPL)3.05 | 10 | 16d ago | ||
| Harmful prompts (evaluated on 3 LLMs and 4 guard LLMs) | ArtPrompt | Mean Perplexity3.23 | 10 | 16d ago | |
| Medium Web Browser | WebTrap | Dual-Goal Success Rate23.81 | 7 | 22d ago | |
| Long Web Browser | WebTrap | Dual-Goal Success Rate47.62 | 7 | 22d ago | |
| MetaQA | AURA | Detected Samples8,321 | 3 | 3mo ago | |
| ImageNet | INACTIVE | SSIM0.9867 | 2 | 3mo ago |