# Narrative Generation on Narrative Constraints Dataset

Source: [Narrative Generation on Narrative Constraints Dataset](<https://www.sota2.com/research/sota/narrative-generation-on-narrative-constraints-dataset>)

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

Data updated: 2026-05-12T09:40:57.019Z

Evaluation of models on the Narrative Constraints Dataset focusing on constraint satisfaction and lexical diversity, measured using metrics like Jaccard Similarity, Effective Number, and Unique Constraints.

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 | Narrative Generation |
| Dataset (including any stated split) | Narrative Constraints Dataset |
| Metric | Jaccard Similarity |
| 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

- [Jaccard Similarity](<https://www.sota2.com/research/sota/narrative-generation-on-narrative-constraints-dataset.md?metric=Jaccard%20Similarity>): Higher is better; unit: PERCENT.

- [Effective Number](<https://www.sota2.com/research/sota/narrative-generation-on-narrative-constraints-dataset.md?metric=Effective%20Number>): Higher is better; unit: SCALAR.

- [Unique Constraints](<https://www.sota2.com/research/sota/narrative-generation-on-narrative-constraints-dataset.md?metric=Unique%20Constraints>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Jaccard Similarity | Result date | Report paper |
| --- | --- | --- | --- |
| gpt5 | 31.69 | Not recorded | 6a02b438dbc836b87d824867 |
| gpt5 | 30.7 | Not recorded | 6a02b438dbc836b87d824867 |
| gemini | 25.97 | Not recorded | 6a02b438dbc836b87d824867 |
| gpt5 | 25.46 | Not recorded | 6a02b438dbc836b87d824867 |
| gpt4.1 | 24.58 | Not recorded | 6a02b438dbc836b87d824867 |
| o4mini | 24.2 | Not recorded | 6a02b438dbc836b87d824867 |
| o4mini | 22.32 | Not recorded | 6a02b438dbc836b87d824867 |
| claude | 21.66 | Not recorded | 6a02b438dbc836b87d824867 |
| claude | 21.6 | Not recorded | 6a02b438dbc836b87d824867 |
| gemini | 20.06 | Not recorded | 6a02b438dbc836b87d824867 |
| o4mini | 20.04 | Not recorded | 6a02b438dbc836b87d824867 |
| gemini | 19.36 | Not recorded | 6a02b438dbc836b87d824867 |
| claude | 18.6 | Not recorded | 6a02b438dbc836b87d824867 |
| gpt4.1 | 14.83 | Not recorded | 6a02b438dbc836b87d824867 |
| gpt4.1 | 13.44 | Not recorded | 6a02b438dbc836b87d824867 |
| qwen | 8.96 | Not recorded | 6a02b438dbc836b87d824867 |
| qwen | 8.67 | Not recorded | 6a02b438dbc836b87d824867 |
| qwen | 8.2 | Not recorded | 6a02b438dbc836b87d824867 |

### gpt5

Jaccard Similarity: 31.69. Result ID: 6a02f4a147f34ce924e612ac.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Quality-focused"}. This record does not assert independent reproduction.

### gpt5

Jaccard Similarity: 30.7. Result ID: 6a02f4a147f34ce924e612ad.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Basic"}. This record does not assert independent reproduction.

### gemini

Jaccard Similarity: 25.97. Result ID: 6a02f4a147f34ce924e612ae.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Creativity-focused"}. This record does not assert independent reproduction.

### gpt5

Jaccard Similarity: 25.46. Result ID: 6a02f4a147f34ce924e612af.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Creativity-focused"}. This record does not assert independent reproduction.

### gpt4.1

Jaccard Similarity: 24.58. Result ID: 6a02f4a147f34ce924e612b0.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Creativity-focused"}. This record does not assert independent reproduction.

### o4mini

Jaccard Similarity: 24.2. Result ID: 6a02f4a147f34ce924e612b1.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Quality-focused"}. This record does not assert independent reproduction.

### o4mini

Jaccard Similarity: 22.32. Result ID: 6a02f4a147f34ce924e612b2.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Basic"}. This record does not assert independent reproduction.

### claude

Jaccard Similarity: 21.66. Result ID: 6a02f4a147f34ce924e612b3.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Creativity-focused"}. This record does not assert independent reproduction.

### claude

Jaccard Similarity: 21.6. Result ID: 6a02f4a147f34ce924e612b4.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Quality-focused"}. This record does not assert independent reproduction.

### gemini

Jaccard Similarity: 20.06. Result ID: 6a02f4a147f34ce924e612b5.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Basic"}. This record does not assert independent reproduction.

### o4mini

Jaccard Similarity: 20.04. Result ID: 6a02f4a147f34ce924e612b6.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Creativity-focused"}. This record does not assert independent reproduction.

### gemini

Jaccard Similarity: 19.36. Result ID: 6a02f4a147f34ce924e612b7.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Quality-focused"}. This record does not assert independent reproduction.

### claude

Jaccard Similarity: 18.6. Result ID: 6a02f4a147f34ce924e612b8.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Basic"}. This record does not assert independent reproduction.

### gpt4.1

Jaccard Similarity: 14.83. Result ID: 6a02f4a147f34ce924e612b9.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Quality-focused"}. This record does not assert independent reproduction.

### gpt4.1

Jaccard Similarity: 13.44. Result ID: 6a02f4a147f34ce924e612ba.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Basic"}. This record does not assert independent reproduction.

### qwen

Jaccard Similarity: 8.96. Result ID: 6a02f4a147f34ce924e612bb.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Quality-focused"}. This record does not assert independent reproduction.

### qwen

Jaccard Similarity: 8.67. Result ID: 6a02f4a147f34ce924e612bc.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Creativity-focused"}. This record does not assert independent reproduction.

### qwen

Jaccard Similarity: 8.2. Result ID: 6a02f4a147f34ce924e612bd.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b438dbc836b87d824867.md>)

Recorded evaluation configuration: {"Instruction Type":"Basic"}. This record does not assert independent reproduction.

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