Long-horizon legal inquiry on Taiwanese criminal case Simple (Case A)
97.2CompletenessSoft-FSM
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Soft-FSMvariant=V32026.02 | 97.2 | 0 | 8 | |
| Pure LLMvariant=V12026.02 | 67.4 | 6.4 | 22.5 | |
| Stage Promptvariant=V22026.02 | 33.5 | 38.8 | 30.5 |
| Method | Links | |||
|---|---|---|---|---|
| Soft-FSMvariant=V32026.02 | 97.2 | 0 | 8 | |
| Pure LLMvariant=V12026.02 | 67.4 | 6.4 | 22.5 | |
| Stage Promptvariant=V22026.02 | 33.5 | 38.8 | 30.5 |