Retrieval-Augmented Generation (RAG) enhances Large Language Model (LLM)
output by providing prior knowledge as context to input. This is beneficial for
knowledge-intensive and expert reliant tasks, including legal
question-answering, which require evidence to validate generated text outputs.
We highlight that Case-Based Reasoning (CBR) presents key opportunities to
structure retrieval as part of the RAG process in an LLM. We introduce CBR-RAG,
where CBR cycle's initial retrieval stage, its indexing vocabulary, and
similarity knowledge containers are used to enhance LLM queries with
contextually relevant cases. This integration augments the original LLM query,
providing a richer prompt. We present an evaluation of CBR-RAG, and examine
different representations (i.e. general and domain-specific embeddings) and
methods of comparison (i.e. inter, intra and hybrid similarity) on the task of
legal question-answering. Our results indicate that the context provided by
CBR's case reuse enforces similarity between relevant components of the
questions and the evidence base leading to significant improvements in the
quality of generated answers.

使用检索增强生成（RAG）方法，结合案例推理（CBR）来增强大型语言模型（LLM）在法律问答等知识密集型和专家依赖性任务中的输出，并通过上下文相关的案例提供更丰富的提示，显著提高生成答案的质量。

CBR-RAG：用于 LLMs 中的检索增强生成的基于案例推理的法律问答系统

CBR-RAG: Case-Based Reasoning for Retrieval Augmented Generation in LLMs  for Legal Question Answering

Legal question-answering (QA) systems have the potential to revolutionize the
way legal professionals interact with case law documents. This paper conducts a
comparative analysis of existing artificial intelligence models for their
utility in answering legal questions within the Indian legal system,
specifically focusing on Indian Legal Question Answering (AILQA) and our study
investigates the efficacy of different retrieval and QA algorithms currently
available. Utilizing the OpenAI GPT model as a benchmark, along with query
prompts, our investigation shows that existing AILQA systems can automatically
interpret natural language queries from users and generate highly accurate
responses. This research is particularly focused on applications within the
Indian criminal justice domain, which has its own set of challenges due to its
complexity and resource constraints. In order to rigorously assess the
performance of these models, empirical evaluations are complemented by feedback
from practicing legal professionals, thereby offering a multifaceted view on
the capabilities and limitations of AI in the context of Indian legal
question-answering.

使用 OpenAI GPT 模型作为基准，结合查询提示，本研究比较分析了现有的人工智能模型在印度法律问答系统中回答法律问题方面的实用性，调查了不同的检索和问答算法的效果。该研究重点关注印度刑事司法领域的应用，该领域面临复杂性和资源限制的挑战。通过从实践法律专业人士获得反馈，以严格评估这些模型的性能，从而全面评估了 AI 在印度法律问答背景下的能力和局限性。