# Unilaw-R1: A Large Language Model for Legal Reasoning with Reinforcement Learning and Iterative Inference

arXiv: https://arxiv.org/abs/2510.10072 · HTML: https://mlsanity.com/paper/2510.10072/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 11 of its 13 references that are in the map.
- First posted on arXiv 11 Oct 2025.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)

## Builds on (most cited first, 11 of 11)

- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Large Language Models: A Survey](https://mlsanity.com/paper/2402.06196/) (2024)
- [MinerU: An Open-Source Solution for Precise Document Content Extraction](https://mlsanity.com/paper/2409.18839/) (2024)
- [Marco-o1: Towards Open Reasoning Models for Open-Ended Solutions](https://mlsanity.com/paper/2411.14405/) (2024)
- [DISC-LawLLM: Fine-tuning Large Language Models for Intelligent Legal Services](https://mlsanity.com/paper/2309.11325/) (2023)
- [Fin-R1: A Large Language Model for Financial Reasoning through Reinforcement Learning](https://mlsanity.com/paper/2503.16252/) (2025)
- [Alpha-GPT: Human-AI Interactive Alpha Mining for Quantitative Investment](https://mlsanity.com/paper/2308.00016/) (2023)
- [LLMs Provide Unstable Answers to Legal Questions](https://mlsanity.com/paper/2502.05196/) (2025)
- [Investigating the Shortcomings of LLMs in Step-by-Step Legal Reasoning](https://mlsanity.com/paper/2502.05675/) (2025)
- [Evaluating Test-Time Scaling LLMs for Legal Reasoning: OpenAI o1, DeepSeek-R1, and Beyond](https://mlsanity.com/paper/2503.16040/) (2025)
- [Tasks and Roles in Legal AI: Data Curation, Annotation, and Verification](https://mlsanity.com/paper/2504.01349/) (2025)

## Built on by (most cited first, 12 of 12)

- [Can LLMs Time Travel? Enhancing Temporal Consistency in Legal Agentic Search through Reinforcement Learning](https://mlsanity.com/paper/2605.25920/) (2026)
- [Towards Explainable Adjudicative Variance: Quantifying Judicial Discretion via Gated Multi-Task Learning](https://mlsanity.com/paper/2606.27069/) (2026)
- [NyayaMind- A Framework for Transparent Legal Reasoning and Judgment Prediction in the Indian Legal System](https://mlsanity.com/paper/2604.09069/) (2026)
- [PoliLegalLM: A Technical Report on a Large Language Model for Political and Legal Affairs](https://mlsanity.com/paper/2604.17543/) (2026)
- [Reconciling Contradictory Views on the Effectiveness of SFT in LLMs: An Interaction Perspective](https://mlsanity.com/paper/2605.17967/) (2026)
- [From token probabilities to calibrated confidence: An empirical study of mathematical question answering](https://mlsanity.com/paper/2608.07827/) (2026)
- [SoK: From Generation to Consumption of Privacy Documents in Software Systems](https://mlsanity.com/paper/2608.12511/) (2026)
- [LexAgentHallu: A Hierarchical Benchmark for Profiling Hallucinations in Legal Agents](https://mlsanity.com/paper/2609.09754/) (2026)
- [Before the Arrest: Benchmarking LLMs on Criminal Profiling from Incomplete Evidence](https://mlsanity.com/paper/2609.19965/) (2026)
- [LexReward: A Taxonomy-Driven Reward Framework for Legal Language Models](https://mlsanity.com/paper/2609.39071/) (2026)
- [HuatuoGPT-3: RL-Only Domain Adaptation from Base Models](https://mlsanity.com/paper/2610.05966/) (2026)
- [LegalRewardBench: Benchmarking Reward Models for Legal Preference Evaluation](https://mlsanity.com/paper/2610.18299/) (2026)

## Abstract (arXiv)

Reasoning-focused large language models (LLMs) are rapidly evolving across various domains, yet their capabilities in handling complex legal problems remains underexplored. In this paper, we introduce Unilaw-R1, a large language model tailored for legal reasoning. With a lightweight 7-billion parameter scale, Unilaw-R1 significantly reduces deployment cost while effectively tackling three core challenges in the legal domain: insufficient legal knowledge, unreliable reasoning logic, and weak business generalization. To address these issues, we first construct Unilaw-R1-Data, a high-quality dataset containing 17K distilled and screened chain-of-thought (CoT) samples. Based on this, we adopt a two-stage training strategy combining Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL), which significantly boosts the performance on complex legal reasoning tasks and supports interpretable decision-making in legal AI applications. To assess legal reasoning ability, we also introduce Unilaw-R1-Eval, a dedicated benchmark designed to evaluate models across single- and multi-choice legal tasks. Unilaw-R1 demonstrates strong results on authoritative benchmarks, outperforming all models of similar scale and achieving performance on par with the much larger DeepSeek-R1-Distill-Qwen-32B (54.9%). Following domain-specific training, it also showed significant gains on LawBench and LexEval, exceeding Qwen-2.5-7B-Instruct (46.6%) by an average margin of 6.6%.