# Train, Retrieve, or Both? A Four-Arm Head-to-Head for Correct Statutory Citation on the Ontario Residential Tenancies Act

arXiv: https://arxiv.org/abs/2606.20359 · HTML: https://mlsanity.com/paper/2606.20359/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 18 references that are in the map.
- First posted on arXiv 18 Jun 2026.

## Builds on (most cited first, 17 of 17)

- [SimRAG: Self-Improving Retrieval-Augmented Generation for Adapting Large Language Models to Specialized Domains](https://mlsanity.com/paper/2410.17952/) (2024)
- [CaLM: Contrasting Large and Small Language Models to Verify Grounded Generation](https://mlsanity.com/paper/2406.05365/) (2024)
- [LawPal : A Retrieval Augmented Generation Based System for Enhanced Legal Accessibility in India](https://mlsanity.com/paper/2502.16573/) (2025)
- [NyayaRAG: Realistic Legal Judgment Prediction with RAG under the Indian Common Law System](https://mlsanity.com/paper/2508.00709/) (2025)
- [LegalRAG: A Hybrid RAG System for Multilingual Legal Information Retrieval](https://mlsanity.com/paper/2504.16121/) (2025)
- [IL-PCSR: Legal Corpus for Prior Case and Statute Retrieval](https://mlsanity.com/paper/2511.00268/) (2025)
- [How Much is Too Much? Exploring LoRA Rank Trade-offs for Retaining Knowledge and Domain Robustness](https://mlsanity.com/paper/2512.15634/) (2025)
- [TRACE: TRansformer-based Attribution using Contrastive Embeddings in LLMs](https://mlsanity.com/paper/2407.04981/) (2024)
- [Learning Beyond the Surface: How Far Can Continual Pre-Training with LoRA Enhance LLMs' Domain-Specific Insight Learning?](https://mlsanity.com/paper/2501.17840/) (2025)
- [Cite Pretrain: Retrieval-Free Knowledge Attribution for Large Language Models](https://mlsanity.com/paper/2506.17585/) (2025)
- [Citation-Grounded Code Comprehension: Preventing LLM Hallucination Through Hybrid Retrieval and Graph-Augmented Context](https://mlsanity.com/paper/2512.12117/) (2025)
- [Judgement Citation Retrieval using Contextual Similarity](https://mlsanity.com/paper/2406.01609/) (2024)
- [Segment First, Retrieve Better: Realistic Legal Search via Rhetorical Role-Based Queries](https://mlsanity.com/paper/2508.00679/) (2025)
- [Scaling Legal AI: Benchmarking Mamba and Transformers for Statutory Classification and Case Law Retrieval](https://mlsanity.com/paper/2509.00141/) (2025)
- [Attribution-Guided Decoding](https://mlsanity.com/paper/2509.26307/) (2025)
- [HalluGraph: Auditable Hallucination Detection for Legal RAG Systems via Knowledge Graph Alignment](https://mlsanity.com/paper/2512.01659/) (2025)
- [Suppressing Domain-Specific Hallucination in Construction LLMs: A Knowledge Graph Foundation for GraphRAG and QLoRA on River and Sediment Control Technical Standards](https://mlsanity.com/paper/2603.13307/) (2026)

## Abstract (arXiv)

Self-represented tenants, landlords, and help-desk staff need to be pointed at the provision of law that actually governs a question, with a correct statutory citation. We study this task on the Ontario Residential Tenancies Act, 2006 (RTA) and its core regulation, asking the operator's question empirically: is fine-tuning enough, or is hybrid retrieval needed? We run a four-arm head-to-head on Qwen2.5-7B-Instruct (base zero-shot, LoRA SFT-only, RAG-only, and an SFT+RAG hybrid), scored on citation exact-match (section+subsection) over a small, human-verification-pending real eval set. The base model cannot cite the RTA and SFT-only mis-recalls sections; retrieval is essential and drives hallucination to zero by construction; and the SFT+RAG hybrid scores highest at 0.481 exact-match with zero hallucinated citations. Its edge comes from SFT making provision selection more robust to the higher-recall candidate sets that hurt zero-shot RAG. Notably, this cheap bge-small hybrid matches or beats a pipeline built on bigger, specialized retrieval models (a larger embedder and a cross-encoder reranker), and a larger/improved training set does not help either: strong statutory-citation performance here does not require specialized retrieval models or more data. The artifact zeroes hallucination and clears the lift-over-base bar but does not reach the aspirational 0.70 exact-match target. All results are on a small, human-verification-pending real eval set and are reported as preliminary.