Multi-step Reasoning on MUSR Murder
68AccuracyStatic Workflow
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Static WorkflowWorkflow Type=Static, LLM Model=DeepSeek-V3.12026.05 | 68 | 0.47 | |
| Dynamic Workflow (ReAct)Workflow Type=Dynamic, LLM Model=DeepSeek-V3.12026.05 | 62 | 1.34 | |
| Zero-shot (Default Imp.)Workflow Type=Zero-shot, LLM Model=DeepSeek-V3.12026.05 | 52 | 0.11 |