| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Multi-turn Agentic Tool-use | TauBench-Airline out-of-distribution | Success Rate (SR)35.5 | 10 | |
| Multi-turn Agentic Tool-use | TauBench Retail | Success Rate (SR)50.3 | 10 | |
| Agentic Reasoning | TauBench V2 | Airline Score76 | 9 | |
| Customer Service | TauBench-Telephony (TBTel) | Accuracy90.5 | 8 | |
| Customer Service | TauBench (TauB) | Accuracy91.8 | 8 | |
| Domain-specific agentic reasoning | TauBench V3 | Airline Accuracy85.8 | 7 | |
| Agentic | TauBench V2 | Airline Score56.9 | 5 | |
| Agentic Reasoning | TauBench Telecom V2 | Accuracy42.7 | 2 |