Password Cracking on RockYou
6.89Mean CRExp. 3 (Ensemble)
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Exp. 3 (Ensemble)n=100, LLM (generation)=qwen3-235b (0.5), gemini-3-flash (0.5), Evaluator LLM=qwen3-max2026.04 | 6.89 | 1.48 | 1.07 | 8.48 | 4.2 | |
| Exp. 1 (Ollama)n=100, LLM (generation)=qwen3:8b, Evaluator LLM=qwen3:8b2026.04 | 4.13 | 3.51 | 0 | 8.28 | 4.1 | |
| Exp. 2 (Gemini)n=100, LLM (generation)=gemini-2.5-flash-lite, Evaluator LLM=qwen3-max2026.04 | 3.79 | 3.15 | 0 | 7.58 | 3.75 | |
| Baseline2026.04 | 2.02 | — | — | 2.02 | — |