# Legal Question Answering on LEXam Public Law (n=12)

Source: [Legal Question Answering on LEXam Public Law (n=12)](<https://www.sota2.com/research/sota/legal-question-answering-on-lexam-public-law-n-12>)

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

Data updated: 2026-06-08T07:12:17.066Z

Evaluation of model performance on the specific n=12 subset of the LEXam Public Law dataset for the Legal Question Answering task, measured using Accuracy and Accuracy Gain.

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 Question Answering |
| Dataset (including any stated split) | LEXam Public Law (n=12) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/legal-question-answering-on-lexam-public-law-n-12.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Accuracy Gain](<https://www.sota2.com/research/sota/legal-question-answering-on-lexam-public-law-n-12.md?metric=Accuracy%20Gain>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| ZS (no traces/HUBO) | 83.3 | Not recorded | 6a264c5790966ade919be7de |
| EP-HUBO (Qwen-35B) | 66.7 | Not recorded | 6a264c5790966ade919be7de |
| EP-HUBO (OSS-20B) | 66.7 | Not recorded | 6a264c5790966ade919be7de |
| MV (Qwen-35B) | 58.3 | Not recorded | 6a264c5790966ade919be7de |
| ZS (no traces/HUBO) | 50 | Not recorded | 6a264c5790966ade919be7de |
| EP-HUBO (Qwen-35B) | 50 | Not recorded | 6a264c5790966ade919be7de |
| EP-HUBO (OSS-20B) | 33.3 | Not recorded | 6a264c5790966ade919be7de |
| MV (OSS-20B) | 25 | Not recorded | 6a264c5790966ade919be7de |

### ZS (no traces/HUBO)

Accuracy: 83.3. Result ID: 6a266a20fbc373592c22ae68.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7de.md>)

Recorded evaluation configuration: {"Adjudicator":"Opus 4.6"}. This record does not assert independent reproduction.

### EP-HUBO (Qwen-35B)

Accuracy: 66.7. Result ID: 6a266a20fbc373592c22ae6b.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7de.md>)

Recorded evaluation configuration: {"Adjudicator":"Opus 4.6","Trace Generator":"Qwen3.5-35B"}. This record does not assert independent reproduction.

### EP-HUBO (OSS-20B)

Accuracy: 66.7. Result ID: 6a266a20fbc373592c22ae6e.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7de.md>)

Recorded evaluation configuration: {"Adjudicator":"Opus 4.6","Trace Generator":"OSS-20B"}. This record does not assert independent reproduction.

### MV (Qwen-35B)

Accuracy: 58.3. Result ID: 6a266a20fbc373592c22ae69.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb61e209131cea41b33.md>)

Recorded evaluation configuration: {"Trace Generator":"Qwen3.5-35B"}. This record does not assert independent reproduction.

### ZS (no traces/HUBO)

Accuracy: 50. Result ID: 6a266a20fbc373592c22ae67.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7de.md>)

Recorded evaluation configuration: {"Adjudicator":"Sonnet 4.6"}. This record does not assert independent reproduction.

### EP-HUBO (Qwen-35B)

Accuracy: 50. Result ID: 6a266a20fbc373592c22ae6a.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7de.md>)

Recorded evaluation configuration: {"Adjudicator":"Sonnet 4.6","Trace Generator":"Qwen3.5-35B"}. This record does not assert independent reproduction.

### EP-HUBO (OSS-20B)

Accuracy: 33.3. Result ID: 6a266a20fbc373592c22ae6d.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7de.md>)

Recorded evaluation configuration: {"Adjudicator":"Sonnet 4.6","Trace Generator":"OSS-20B"}. This record does not assert independent reproduction.

### MV (OSS-20B)

Accuracy: 25. Result ID: 6a266a20fbc373592c22ae6c.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb61e209131cea41b33.md>)

Recorded evaluation configuration: {"Trace Generator":"OSS-20B"}. This record does not assert independent reproduction.

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