LeMAJ (Legal LLM-as-a-Judge): Bridging Legal Reasoning and LLM Evaluation
This addresses the problem of reliable LLM evaluation for legal applications, which is incremental as it builds on the LLM-as-a-Judge approach but tailors it to legal reasoning.
The paper tackles the challenge of evaluating large language model outputs in the legal domain by introducing a reference-free method that breaks down responses into Legal Data Points, outperforming baselines on proprietary and open-source datasets and correlating more closely with human expert evaluations.
Evaluating large language model (LLM) outputs in the legal domain presents unique challenges due to the complex and nuanced nature of legal analysis. Current evaluation approaches either depend on reference data, which is costly to produce, or use standardized assessment methods, both of which have significant limitations for legal applications. Although LLM-as-a-Judge has emerged as a promising evaluation technique, its reliability and effectiveness in legal contexts depend heavily on evaluation processes unique to the legal industry and how trustworthy the evaluation appears to the human legal expert. This is where existing evaluation methods currently fail and exhibit considerable variability. This paper aims to close the gap: a) we break down lengthy responses into 'Legal Data Points' (LDPs), self-contained units of information, and introduce a novel, reference-free evaluation methodology that reflects how lawyers evaluate legal answers; b) we demonstrate that our method outperforms a variety of baselines on both our proprietary dataset and an open-source dataset (LegalBench); c) we show how our method correlates more closely with human expert evaluations and helps improve inter-annotator agreement; and finally d) we open source our Legal Data Points for a subset of LegalBench used in our experiments, allowing the research community to replicate our results and advance research in this vital area of LLM evaluation on legal question-answering.