English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Large language models (LLMs) have demonstrated great potential fordomain-specific applications, such as the law domain. However, recent disputesover GPT-4's law evaluation raise questions concerning their performance inreal-world legal tasks. To systematically investigate their competency in thelaw, we design practical baseline solutions based on LLMs and test on the taskof legal judgment prediction. In our solutions, LLMs can work alone to answeropen questions or coordinate with an information retrieval (IR) system to learnfrom similar cases or solve simplified multi-choice questions. We show thatsimilar cases and multi-choice options, namely label candidates, included inprompts can help LLMs recall domain knowledge that is critical for expertiselegal reasoning. We additionally present an intriguing paradox wherein an IRsystem surpasses the performance of LLM+IR due to limited gains acquired byweaker LLMs from powerful IR systems. In such cases, the role of LLMs becomesredundant. Our evaluation pipeline can be easily extended into other tasks tofacilitate evaluations in other domains. Code is available athttps://github.com/srhthu/LM-CompEval-Legal

A Comprehensive Evaluation of Large Language Models on Legal Judgment Prediction