# Creative Writing on ROCStories + WritingPrompts Llama-3-70B (test)

Source: [Creative Writing on ROCStories + WritingPrompts Llama-3-70B (test)](<https://www.sota2.com/research/sota/creative-writing-on-rocstories-plus-writingprompts-llama-3-70b-test>)

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

Data updated: 2026-05-12T13:02:05.219Z

Evaluation of creative writing performance on a combined test set derived from ROCStories and WritingPrompts, specifically evaluated under the Llama-3-70B configuration, using metrics such as QD-Score, Covariance, and SB.

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 | Creative writing |
| Dataset (including any stated split) | ROCStories + WritingPrompts Llama-3-70B (test) |
| Metric | QD-Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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

- [QD-Score](<https://www.sota2.com/research/sota/creative-writing-on-rocstories-plus-writingprompts-llama-3-70b-test.md?metric=QD-Score>): Higher is better; unit: SCALAR.

- [Median Score \[IQR\]](<https://www.sota2.com/research/sota/creative-writing-on-rocstories-plus-writingprompts-llama-3-70b-test.md?metric=Median%20Score%20%5BIQR%5D>): Higher is better; unit: SCALAR.

- [Covariance](<https://www.sota2.com/research/sota/creative-writing-on-rocstories-plus-writingprompts-llama-3-70b-test.md?metric=Covariance>): Higher is better; unit: SCALAR.

- [SB](<https://www.sota2.com/research/sota/creative-writing-on-rocstories-plus-writingprompts-llama-3-70b-test.md?metric=SB>): Lower is better; unit: SCALAR.

- [Metric A](<https://www.sota2.com/research/sota/creative-writing-on-rocstories-plus-writingprompts-llama-3-70b-test.md?metric=Metric%20A>): Higher is better; unit: SCALAR.

## Recorded results

| Method | QD-Score | Result date | Report paper |
| --- | --- | --- | --- |
| QD-LLM | 63.3 | Not recorded | 6a02b449dbc836b87d824965 |
| CMA-ME (ad.) | 46.2 | Not recorded | 6a02b449dbc836b87d824965 |
| QDAIF | 44.8 | Not recorded | 6a02b449dbc836b87d824965 |
| Best-of-N+MMR | 40.1 | Not recorded | 6a02b449dbc836b87d824965 |
| Diverse Beam | 38.2 | Not recorded | 6a02b449dbc836b87d824965 |
| Nucleus | 36.4 | Not recorded | 6a02b449dbc836b87d824965 |

### QD-LLM

QD-Score: 63.3. Result ID: 6a0323e747f34ce924e77882.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b449dbc836b87d824965.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b449dbc836b87d824965.md>)

Recorded evaluation configuration: {"Backbone":"Llama-3-70B-Instruct","Budget":"500"}. This record does not assert independent reproduction.

### CMA-ME (ad.)

QD-Score: 46.2. Result ID: 6a0323e747f34ce924e77880.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b449dbc836b87d824965.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67741e209131cea1669b.md>)

Recorded evaluation configuration: {"Backbone":"Llama-3-70B-Instruct","Budget":"500"}. This record does not assert independent reproduction.

### QDAIF

QD-Score: 44.8. Result ID: 6a0323e747f34ce924e77881.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b449dbc836b87d824965.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71661e209131cea51324.md>)

Recorded evaluation configuration: {"Backbone":"Llama-3-70B-Instruct","Budget":"500"}. This record does not assert independent reproduction.

### Best-of-N+MMR

QD-Score: 40.1. Result ID: 6a0323e747f34ce924e7787f.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b449dbc836b87d824965.md>)

Recorded evaluation configuration: {"Backbone":"Llama-3-70B-Instruct","Budget":"500"}. This record does not assert independent reproduction.

### Diverse Beam

QD-Score: 38.2. Result ID: 6a0323e747f34ce924e7787e.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b449dbc836b87d824965.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64fd1e209131cea07a80.md>)

Recorded evaluation configuration: {"Backbone":"Llama-3-70B-Instruct","Budget":"500"}. This record does not assert independent reproduction.

### Nucleus

QD-Score: 36.4. Result ID: 6a0323e747f34ce924e7787d.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b449dbc836b87d824965.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67041e209131cea13c72.md>)

Recorded evaluation configuration: {"Backbone":"Llama-3-70B-Instruct","p":"0.95","Budget":"500"}. This record does not assert independent reproduction.

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