| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Legal Reasoning | CaseHOLD (test) | Test Accuracy89.22 | 22 | |
| Legal Reasoning | CaseHold | Accuracy (CaseHold)83.13 | 16 | |
| Case holding classification | CaseHOLD (test) | Mean macro F178.5 | 12 | |
| Question Answering | CaseHOLD | AR (%)100 | 9 | |
| Model Stealing Defense | CaseHOLD | Agreement (%)41.5 | 8 | |
| Legal Reasoning | CaseHold | Cumulative Score (CS)96 | 8 | |
| Question Answering | CaseHOLD (eval) | Risk7.8 | 3 | |
| Question Answering | CaseHOLD (CHOLD) | Accuracy/F1 Score74.1 | 2 |