Purpose
LawBench is an evaluation benchmark designed specifically for the legal domain for large language models, with the goal of systematically evaluating the capabilities of large language models in legal scenarios.
Main Capabilities
This project provides an online leaderboard (Leaderboard), used to display the evaluation performance of models on the LawBench benchmark.
How to Use
Its leaderboard page can be accessed through its official website:https://lawbench.opencompass.org.cn/leaderboard . Information on how to submit models, how to reproduce evaluations, whether there is local code or installation commands, etc. is not yet provided in the materials.
Applicable Scenarios
Suitable for scenarios where there is a need to understand or compare the performance of large language models in the legal domain.
Limitations
Regarding the specific types of legal tasks covered by this benchmark, data sources, evaluation metrics, model rankings, whether it is open source, whether it is paid, etc., the materials do not yet provide this, and it cannot be confirmed.
