Multi-step Reasoning on MUSR Objects (Accuracy, Cost)
58AccuracyStatic Workflow
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Static WorkflowWorkflow Type=Static, LLM Model=DeepSeek-V3.12026.05 | 58 | 0.34 | |
| Dynamic Workflow (ReAct)Workflow Type=Dynamic, LLM Model=DeepSeek-V3.12026.05 | 36 | 0.99 | |
| Zero-shot (Default Imp.)Workflow Type=Zero-shot, LLM Model=DeepSeek-V3.12026.05 | 35 | 0.1 |