# Schema Generation Quality Evaluation on Technical Evaluation Test Suite (6 domains)

Source: [Schema Generation Quality Evaluation on Technical Evaluation Test Suite (6 domains)](<https://www.sota2.com/research/sota/schema-generation-quality-evaluation-on-technical-evaluation-test-suite-6-domains>)

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

Data updated: 2026-04-10T06:48:31.988Z

Evaluation of model generated schemas across six distinct content domains (including job abstracts, videos, and pitch decks), assessed using metrics related to schema quality, example fit, and generalizability.

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 | Schema Generation Quality Evaluation |
| Dataset (including any stated split) | Technical Evaluation Test Suite 6 domain topics: job talk abstracts, teaser videos, LinkedIn summaries, pitch decks, news headlines, news TikToks |
| Metric | Schema Granularity |
| Unit | SCALAR |
| 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

- [Schema Granularity](<https://www.sota2.com/research/sota/schema-generation-quality-evaluation-on-technical-evaluation-test-suite-6-domains.md?metric=Schema%20Granularity>): Higher is better; unit: SCALAR.

- [Example Fit](<https://www.sota2.com/research/sota/schema-generation-quality-evaluation-on-technical-evaluation-test-suite-6-domains.md?metric=Example%20Fit>): Higher is better; unit: SCALAR.

- [Generation Quality](<https://www.sota2.com/research/sota/schema-generation-quality-evaluation-on-technical-evaluation-test-suite-6-domains.md?metric=Generation%20Quality>): Higher is better; unit: SCALAR.

- [Generalizability](<https://www.sota2.com/research/sota/schema-generation-quality-evaluation-on-technical-evaluation-test-suite-6-domains.md?metric=Generalizability>): Higher is better; unit: SCALAR.

- [P-value (1 vs 2)](<https://www.sota2.com/research/sota/schema-generation-quality-evaluation-on-technical-evaluation-test-suite-6-domains.md?metric=P-value%20(1%20vs%202)>): Lower is better; unit: SCALAR.

- [P-value (2 vs 3)](<https://www.sota2.com/research/sota/schema-generation-quality-evaluation-on-technical-evaluation-test-suite-6-domains.md?metric=P-value%20(2%20vs%203)>): Lower is better; unit: SCALAR.

- [P-value (1 vs 3)](<https://www.sota2.com/research/sota/schema-generation-quality-evaluation-on-technical-evaluation-test-suite-6-domains.md?metric=P-value%20(1%20vs%203)>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Schema Granularity | Result date | Report paper |
| --- | --- | --- | --- |
| Schemex Full | 6.8 | Not recorded | 68fa7b691e209131cea8841d |
| Schemex Initial | 5.9 | Not recorded | 68fa7b691e209131cea8841d |
| o1-pro | 5 | Not recorded | 68fa7b691e209131cea8841d |

### Schemex Full

Schema Granularity: 6.8. Result ID: 69d89c7347f34ce924c0dd8b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea8841d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea8841d.md>)

Recorded evaluation configuration: {"Condition":"3","Description":"full system with one iteration of contrastive refinement building on Schemex Initial","Contrastive refinement iteration":"1"}. This record does not assert independent reproduction.

### Schemex Initial

Schema Granularity: 5.9. Result ID: 69d89c7347f34ce924c0dd8a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea8841d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea8841d.md>)

Recorded evaluation configuration: {"Condition":"2","Description":"structured workflow with clustering followed by abstraction into a schema","Contrastive refinement iteration":"0"}. This record does not assert independent reproduction.

### o1-pro

Schema Granularity: 5. Result ID: 69d89c7347f34ce924c0dd89.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea8841d.md>)

Recorded evaluation configuration: {"Condition":"1","Reasoning model":"state-of-the-art extended reasoning model","Context window":"200K-token"}. This record does not assert independent reproduction.

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