A new research paper explores the effectiveness of different AI approaches for accurately citing legal statutes, specifically the Ontario Residential Tenancies Act. The study compared a base model, a fine-tuned model, a retrieval-augmented generation (RAG) model, and a hybrid SFT+RAG model. Results indicate that retrieval methods are crucial for reducing hallucinations and achieving correct citations, with the hybrid SFT+RAG model achieving the highest exact-match score of 0.481. AI
IMPACT Demonstrates the necessity of retrieval augmentation for accurate legal citation, potentially influencing how AI is used in legal tech.
RANK_REASON The cluster contains an academic paper detailing experimental results on AI model performance for a specific task.
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
