LegalDrill: Diagnosis-Driven Synthesis for Legal Reasoning in Small Language Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918468478042112 |
|---|---|
| author | Li, Tianchun Liu, Haochen Pardeshi, Vishwa Wang, Xingchen Liu, Tianci Zhao, Huijun Fan, Wei Gao, Jing |
| author_facet | Li, Tianchun Liu, Haochen Pardeshi, Vishwa Wang, Xingchen Liu, Tianci Zhao, Huijun Fan, Wei Gao, Jing |
| contents | Small language models (SLMs) are promising for real-world deployment due to their efficiency and low operational cost. However, their limited capacity struggles with high-stakes legal reasoning tasks that require coherent statute interpretation and logically consistent deduction. Furthermore, training SLMs for such tasks demands high-quality, concise reasoning trajectories, which are prohibitively expensive to manually collect and difficult to curate via standard rejection sampling, lacking granularity beyond final verdicts. To address these challenges, we propose {LegalDrill}, a diagnosis-driven synthesis framework that extracts and iteratively refines reasoning trajectories from a capable teacher via fine-grained prompting, then a self-reflective verification is employed to adaptively select the most effective data for the SLM student. The resulting data empower SLM training through supervised fine-tuning and direct preference optimization. Extensive experiments on several legal benchmarks demonstrate that {LegalDrill} significantly bolsters the legal reasoning capabilities of representative SLMs while bypassing the need for scarce expert annotations, paving a scalable path toward practical legal reasoning systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_23809 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | LegalDrill: Diagnosis-Driven Synthesis for Legal Reasoning in Small Language Models Li, Tianchun Liu, Haochen Pardeshi, Vishwa Wang, Xingchen Liu, Tianci Zhao, Huijun Fan, Wei Gao, Jing Computation and Language Small language models (SLMs) are promising for real-world deployment due to their efficiency and low operational cost. However, their limited capacity struggles with high-stakes legal reasoning tasks that require coherent statute interpretation and logically consistent deduction. Furthermore, training SLMs for such tasks demands high-quality, concise reasoning trajectories, which are prohibitively expensive to manually collect and difficult to curate via standard rejection sampling, lacking granularity beyond final verdicts. To address these challenges, we propose {LegalDrill}, a diagnosis-driven synthesis framework that extracts and iteratively refines reasoning trajectories from a capable teacher via fine-grained prompting, then a self-reflective verification is employed to adaptively select the most effective data for the SLM student. The resulting data empower SLM training through supervised fine-tuning and direct preference optimization. Extensive experiments on several legal benchmarks demonstrate that {LegalDrill} significantly bolsters the legal reasoning capabilities of representative SLMs while bypassing the need for scarce expert annotations, paving a scalable path toward practical legal reasoning systems. |
| title | LegalDrill: Diagnosis-Driven Synthesis for Legal Reasoning in Small Language Models |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2604.23809 |