| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Legal Knowledge and Reasoning Benchmark | LawBench | Memorization Score69.19 | 30 | |
| Legal Evaluation | LawBench | Accuracy62 | 26 | |
| Legal Question Answering | LawBench revised (test) | ROUGE-144.68 | 17 | |
| Legal Reasoning | LawBench | Micro-F170 | 10 | |
| Reading Comprehension | LawBench 2-5 | RC-F197.59 | 8 | |
| Legal Charge Prediction | LawBench 8 (test) | F1 Score69.51 | 6 | |
| Hallucination Detection | LawBench Open-ended tasks | Average Entropy1.68 | 5 | |
| Legal Reasoning | LawBench | Top-1 Accuracy70.1 | 4 | |
| Autonomous LLM Fine-tuning | LawBench | Hybrid Score51.2 | 4 | |
| Legal Consultation | LawBench Revised | Legal Soundness4.14 | 3 |