Cross-lingual Question Answering on LEGALBENCH RuleQA English (test)
20.25ROUGE-1JurisMA
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| JurisMAModel category=Our Method, Averaging seeds=52026.04 | 20.25 | 10.72 | 12.81 | 16.94 | 6.72 | 5.41 | 70.48 | |
| GPT4oModel category=General LLM, Averaging seeds=52026.04 | 10.1 | 1.4 | 8.11 | 4.66 | 0.49 | 0.21 | 57.88 | |
| Lawyer-llamaModel category=Legal LLM, Averaging seeds=52026.04 | 8.42 | 0.07 | 7.7 | 3.7 | 0.03 | 0.97 | 57.13 |