# Long-horizon legal inquiry on Taiwanese criminal case Simple (Case A)

Source: [Long-horizon legal inquiry on Taiwanese criminal case Simple (Case A)](<https://www.sota2.com/research/sota/long-horizon-legal-inquiry-on-taiwanese-criminal-case-simple-case-a>)

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

Data updated: 2026-02-26T05:39:38.161Z

Evaluation of model performance on complex, multi-step legal questioning using the Taiwanese criminal case Simple (Case A) dataset, focusing on completeness, redundancy, and unknown rate.

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 | Long-horizon legal inquiry |
| Dataset (including any stated split) | Taiwanese criminal case Simple (Case A) |
| Metric | Completeness |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [Completeness](<https://www.sota2.com/research/sota/long-horizon-legal-inquiry-on-taiwanese-criminal-case-simple-case-a.md?metric=Completeness>): Higher is better; unit: PERCENT.

- [Redundancy](<https://www.sota2.com/research/sota/long-horizon-legal-inquiry-on-taiwanese-criminal-case-simple-case-a.md?metric=Redundancy>): Lower is better; unit: PERCENT.

- [Unknown Rate](<https://www.sota2.com/research/sota/long-horizon-legal-inquiry-on-taiwanese-criminal-case-simple-case-a.md?metric=Unknown%20Rate>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Completeness | Result date | Report paper |
| --- | --- | --- | --- |
| Soft-FSM | 97.2 | Not recorded | 69842469a8b01e76cc8ee968 |
| Pure LLM | 67.4 | Not recorded | 69842469a8b01e76cc8ee968 |
| Stage Prompt | 33.5 | Not recorded | 69842469a8b01e76cc8ee968 |

### Soft-FSM

Completeness: 97.2. Result ID: 69964de75698665ec492733e.

[Reporting paper](<https://www.sota2.com/research/paper/69842469a8b01e76cc8ee968.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69842469a8b01e76cc8ee968.md>)

Recorded evaluation configuration: {"variant":"V3"}. This record does not assert independent reproduction.

### Pure LLM

Completeness: 67.4. Result ID: 69964de75698665ec492733c.

[Reporting paper](<https://www.sota2.com/research/paper/69842469a8b01e76cc8ee968.md>)

Recorded evaluation configuration: {"variant":"V1"}. This record does not assert independent reproduction.

### Stage Prompt

Completeness: 33.5. Result ID: 69964de75698665ec492733d.

[Reporting paper](<https://www.sota2.com/research/paper/69842469a8b01e76cc8ee968.md>)

Recorded evaluation configuration: {"variant":"V2"}. This record does not assert independent reproduction.

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