Tool-augmented Question Answering on Musique
17.33AccuracyEPC-AW
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| EPC-AWBackbone=Qwen3-Coder-30B2026.05 | 17.33 | — | |
| RollbackBackbone=Qwen3-Coder-30B2026.05 | 14 | 3.33 | |
| RetryBackbone=Qwen3-Coder-30B2026.05 | 10.67 | 6.66 | |
| No-Repair (AgentFlow)Backbone=Qwen3-Coder-30B2026.05 | 9.33 | 8 |