overthelex/ua-statute-retrieval

UA-StatuteRetrieval: Legal Article Prediction Benchmark

A large-scale benchmark for evaluating legal statute retrieval methods on Ukrainian court decisions. Ground truth is derived exhaustively from 37M+ codex citations extracted from the Ukrainian Unified State Register of Court Decisions (EDRSR).

Dataset Description

Instead of manual annotations, this benchmark exploits implicit relevance judgments: when a judge cites Article 625 of the Civil Code, that constitutes a relevance judgment for that article given the case context.

Key Statistics (2024)

• 37.5M codex citations extracted
• 2.25M court decisions
• 3,671 statutory articles evaluated
• 1.8M retrieval predictions (leave-one-out protocol)

Files

File	Description
article_retrieval_performance.parquet	Per-article metrics: degree, mean rank, Hit@5, MRR for 3,667 articles (year 2024)
temporal_metrics.parquet	Retrieval metrics by year (2007-2026) showing temporal degradation
temporal_metrics.csv	Same as above in CSV format
difficulty_stratification.csv	Performance by article frequency bin (hub/high/mid/low/rare)

Evaluation Protocol

Leave-one-out on the full bipartite citation graph:

For each court decision with cited articles {a1, ..., an}:

1. Mask each article ai in turn
2. Score all candidate articles using remaining citations as seed
3. Compute rank of masked article among non-seed candidates

Baselines

Method	Hit@10	MRR
Adamic-Adar	0.545	0.272
Common Neighbors	0.534	0.266
Degree Baseline	0.111	0.059

Difficulty Stratification

Bin	Articles	Hit@10 (AA)	MRR (AA)
Hub (>100K cites)	21	0.890	0.536
High (10K-100K)	354	0.601	0.274
Mid (1K-10K)	864	0.138	0.074
Low (100-1K)	1,689	0.041	0.020
Rare (<100)	739	0.012	0.010

Temporal Degradation

MRR degrades from 0.42-0.56 (2008-2012) to 0.27-0.28 (2024-2026), demonstrating that co-citation patterns become less predictive over time as legal practice evolves.

Usage

from datasets import load_dataset

ds = load_dataset("overthelex/ua-statute-retrieval")

# Per-article performance
articles = ds["train"].to_pandas()
print(articles[["target_article", "degree", "mrr_aa", "hit5_aa"]].head(20))

Citation

@article{keshchenko2025statute,
  title={UA-StatuteRetrieval: A Full-Graph Benchmark for Legal Article Prediction from 37M Citation Records},
  author={Keshchenko, Volodymyr},
  year={2025}
}

Source Data

Derived from the Ukrainian Unified State Register of Court Decisions (EDRSR). Raw citation data (502M records) is not published.

Related Datasets

• overthelex/ua-court-citation-graph - Co-citation graph
• overthelex/ukrainian-court-decisions - Court decision metadata
• overthelex/ua-case-outcome - Case outcome prediction