Intent Mismatch Detection on Generative System Failure Modes
0.694Detection AccuracyQQJ
Evaluation Results
| Method | Links | |
|---|---|---|
| QQJ2026.05 | 0.694 | |
| G-Eval2026.05 | 0.527 | |
| LLM Reviewer2026.05 | 0.502 | |
| MT-Bench2026.05 | 0.443 | |
| BLEU / ROUGE2026.05 | 0.197 |
| Method | Links | |
|---|---|---|
| QQJ2026.05 | 0.694 | |
| G-Eval2026.05 | 0.527 | |
| LLM Reviewer2026.05 | 0.502 | |
| MT-Bench2026.05 | 0.443 | |
| BLEU / ROUGE2026.05 | 0.197 |