# Legal Article Retrieval on LexRAG (test)

Source: [Legal Article Retrieval on LexRAG (test)](<https://www.sota2.com/research/sota/legal-article-retrieval-on-lexrag-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-29T11:13:55.282Z

Evaluation of Legal Article Retrieval performance on the test split of the LexRAG dataset, measured using Recall@K and NDCG@K metrics.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Legal Article Retrieval |
| Dataset (including any stated split) | LexRAG (test) |
| Metric | Recall@1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 18 of 18 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Recall@1](<https://www.sota2.com/research/sota/legal-article-retrieval-on-lexrag-test.md?metric=Recall%401>): Higher is better; unit: PERCENT.

- [Recall@3](<https://www.sota2.com/research/sota/legal-article-retrieval-on-lexrag-test.md?metric=Recall%403>): Higher is better; unit: PERCENT.

- [Recall@5](<https://www.sota2.com/research/sota/legal-article-retrieval-on-lexrag-test.md?metric=Recall%405>): Higher is better; unit: PERCENT.

- [NDCG@1](<https://www.sota2.com/research/sota/legal-article-retrieval-on-lexrag-test.md?metric=NDCG%401>): Higher is better; unit: PERCENT.

- [NDCG@3](<https://www.sota2.com/research/sota/legal-article-retrieval-on-lexrag-test.md?metric=NDCG%403>): Higher is better; unit: PERCENT.

- [NDCG@5](<https://www.sota2.com/research/sota/legal-article-retrieval-on-lexrag-test.md?metric=NDCG%405>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Recall@1 | Result date | Report paper |
| --- | --- | --- | --- |
| LEXPATH | 16.86 | Not recorded | 6a191dbadbd7150506f829c4 |
| A-RAG | 11.36 | Not recorded | 6a191dbadbd7150506f829c4 |
| BM25 + bge-m3 + jina-reranker | 10.83 | Not recorded | 6a191dbadbd7150506f829c4 |
| bge-m3 | 10.54 | Not recorded | 6a191dbadbd7150506f829c4 |
| bge-m3 + jina-reranker | 10.53 | Not recorded | 6a191dbadbd7150506f829c4 |
| BM25 + bge-m3 | 10.04 | Not recorded | 6a191dbadbd7150506f829c4 |
| bge-large-zh-v1.5 | 9.78 | Not recorded | 6a191dbadbd7150506f829c4 |
| BM25 + bge-large-zh-v1.5 | 8.61 | Not recorded | 6a191dbadbd7150506f829c4 |
| multilingual-e5-large | 7.23 | Not recorded | 6a191dbadbd7150506f829c4 |
| CRAG | 5.52 | Not recorded | 6a191dbadbd7150506f829c4 |
| text2vec-base-chinese | 5.26 | Not recorded | 6a191dbadbd7150506f829c4 |
| gte-Qwen2-1.5B | 4.98 | Not recorded | 6a191dbadbd7150506f829c4 |
| BM25 | 4.82 | Not recorded | 6a191dbadbd7150506f829c4 |
| BM25 + jina-reranker | 4.81 | Not recorded | 6a191dbadbd7150506f829c4 |
| QL | 4.64 | Not recorded | 6a191dbadbd7150506f829c4 |
| IRCoT | 3.5 | Not recorded | 6a191dbadbd7150506f829c4 |
| SAILER\_zh | 2.76 | Not recorded | 6a191dbadbd7150506f829c4 |
| ChatLaw\_Text2Vec | 1.97 | Not recorded | 6a191dbadbd7150506f829c4 |

### LEXPATH

Recall@1: 16.86. Result ID: 6a197400fbc373592c14aa48.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

Recorded evaluation configuration: {"Model":"LEXPATH"}. This record does not assert independent reproduction.

### A-RAG

Recall@1: 11.36. Result ID: 6a197400fbc373592c14aa47.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6982d2ada8b01e76cc8e95ac.md>)

[Implementation](<https://github.com/Ayanami0730/arag>)

Recorded evaluation configuration: {"Model":"A-RAG"}. This record does not assert independent reproduction.

### BM25 + bge-m3 + jina-reranker

Recall@1: 10.83. Result ID: 6a197400fbc373592c14aa44.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

Recorded evaluation configuration: {"Model":"BM25 + bge-m3 + jina-reranker"}. This record does not assert independent reproduction.

### bge-m3

Recall@1: 10.54. Result ID: 6a1973fffbc373592c14aa3c.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73d91e209131cea5e7ab.md>)

[Implementation](<https://github.com/FlagOpen/FlagEmbedding>)

Recorded evaluation configuration: {"Model":"bge-m3"}. This record does not assert independent reproduction.

### bge-m3 + jina-reranker

Recall@1: 10.53. Result ID: 6a197400fbc373592c14aa43.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

Recorded evaluation configuration: {"Model":"bge-m3 + jina-reranker"}. This record does not assert independent reproduction.

### BM25 + bge-m3

Recall@1: 10.04. Result ID: 6a197400fbc373592c14aa41.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

Recorded evaluation configuration: {"Model":"BM25 + bge-m3"}. This record does not assert independent reproduction.

### bge-large-zh-v1.5

Recall@1: 9.78. Result ID: 6a1973fffbc373592c14aa3a.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74e31e209131cea64745.md>)

Recorded evaluation configuration: {"Model":"bge-large-zh-v1.5"}. This record does not assert independent reproduction.

### BM25 + bge-large-zh-v1.5

Recall@1: 8.61. Result ID: 6a197400fbc373592c14aa40.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

Recorded evaluation configuration: {"Model":"BM25 + bge-large-zh-v1.5"}. This record does not assert independent reproduction.

### multilingual-e5-large

Recall@1: 7.23. Result ID: 6a1973fffbc373592c14aa3d.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72251e209131cea54c85.md>)

Recorded evaluation configuration: {"Model":"multilingual-e5-large"}. This record does not assert independent reproduction.

### CRAG

Recall@1: 5.52. Result ID: 6a197400fbc373592c14aa46.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75111e209131cea65797.md>)

Recorded evaluation configuration: {"Model":"CRAG"}. This record does not assert independent reproduction.

### text2vec-base-chinese

Recall@1: 5.26. Result ID: 6a1973fffbc373592c14aa39.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

Recorded evaluation configuration: {"Model":"text2vec-base-chinese"}. This record does not assert independent reproduction.

### gte-Qwen2-1.5B

Recall@1: 4.98. Result ID: 6a1973fffbc373592c14aa3b.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa703b1e209131cea49c20.md>)

Recorded evaluation configuration: {"Model":"gte-Qwen2-1.5B"}. This record does not assert independent reproduction.

### BM25

Recall@1: 4.82. Result ID: 6a1973fffbc373592c14aa37.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

Recorded evaluation configuration: {"Model":"BM25"}. This record does not assert independent reproduction.

### BM25 + jina-reranker

Recall@1: 4.81. Result ID: 6a197400fbc373592c14aa42.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

Recorded evaluation configuration: {"Model":"BM25 + jina-reranker"}. This record does not assert independent reproduction.

### QL

Recall@1: 4.64. Result ID: 6a1973fffbc373592c14aa38.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

Recorded evaluation configuration: {"Model":"QL"}. This record does not assert independent reproduction.

### IRCoT

Recall@1: 3.5. Result ID: 6a197400fbc373592c14aa45.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fd81e209131cea4791d.md>)

[Implementation](<https://github.com/stonybrooknlp/ircot>)

Recorded evaluation configuration: {"Model":"IRCoT"}. This record does not assert independent reproduction.

### SAILER\_zh

Recall@1: 2.76. Result ID: 6a197400fbc373592c14aa3f.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f231e209131cea43f91.md>)

Recorded evaluation configuration: {"Model":"SAILER\_zh"}. This record does not assert independent reproduction.

### ChatLaw\_Text2Vec

Recall@1: 1.97. Result ID: 6a1973fffbc373592c14aa3e.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dbadbd7150506f829c4.md>)

Recorded evaluation configuration: {"Model":"ChatLaw\_Text2Vec"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
