# Private Information Tagging on Legal (test)

Source: [Private Information Tagging on Legal (test)](<https://www.sota2.com/research/sota/private-information-tagging-on-legal-test>)

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

Data updated: 2026-03-18T07:04:32.317Z

Evaluation of model performance on tagging private information within the test split of the Legal dataset, measured using Precision, Recall, and F1 scores.

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 | Private Information Tagging |
| Dataset (including any stated split) | Legal (test) |
| Metric | Precision |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Precision](<https://www.sota2.com/research/sota/private-information-tagging-on-legal-test.md?metric=Precision>): Higher is better; unit: PERCENT.

- [Recall](<https://www.sota2.com/research/sota/private-information-tagging-on-legal-test.md?metric=Recall>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/private-information-tagging-on-legal-test.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Precision | Result date | Report paper |
| --- | --- | --- | --- |
| SFT | 78.72 | Not recorded | 69ba317b0636e5bf4afc072c |
| MAC+ | 75.93 | Not recorded | 69ba317b0636e5bf4afc072c |
| GRPO | 64.61 | Not recorded | 69ba317b0636e5bf4afc072c |
| MAC | 64.32 | Not recorded | 69ba317b0636e5bf4afc072c |

### SFT

Precision: 78.72. Result ID: 69ba4e6847f34ce924a3a7b6.

[Reporting paper](<https://www.sota2.com/research/paper/69ba317b0636e5bf4afc072c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b121e209131cea2bf59.md>)

[Implementation](<https://github.com/microsoft/LoRA>)

Recorded evaluation configuration: {"Model Size":"3B","Category":"Training Baseline"}. This record does not assert independent reproduction.

### MAC+

Precision: 75.93. Result ID: 69ba4e6847f34ce924a3a7b8.

[Reporting paper](<https://www.sota2.com/research/paper/69ba317b0636e5bf4afc072c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ba317b0636e5bf4afc072c.md>)

Recorded evaluation configuration: {"Model Size":"3B","Category":"MAC Variant","Tuning Dataset":"Healthcare & Finance"}. This record does not assert independent reproduction.

### GRPO

Precision: 64.61. Result ID: 69ba4e6847f34ce924a3a7b5.

[Reporting paper](<https://www.sota2.com/research/paper/69ba317b0636e5bf4afc072c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

[Implementation](<https://github.com/deepseek-ai/DeepSeek-Math>)

Recorded evaluation configuration: {"Model Size":"3B","Category":"Training Baseline"}. This record does not assert independent reproduction.

### MAC

Precision: 64.32. Result ID: 69ba4e6847f34ce924a3a7b7.

[Reporting paper](<https://www.sota2.com/research/paper/69ba317b0636e5bf4afc072c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ba317b0636e5bf4afc072c.md>)

Recorded evaluation configuration: {"Model Size":"3B","Category":"MAC Variant"}. This record does not assert independent reproduction.

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