# Style Customization on 20 Users OPT-6.7b based (ID)

Source: [Style Customization on 20 Users OPT-6.7b based (ID)](<https://www.sota2.com/research/sota/style-customization-on-20-users-opt-6-7b-based-id>)

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

Data updated: 2026-02-26T05:55:09.823Z

Evaluation of Style Customization performance using metrics such as Perplexity (PPL) and Accuracy, based on data generated or sourced from 20 users interacting with the OPT-6.7b model.

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 | Style Customization |
| Dataset (including any stated split) | 20 Users OPT-6.7b based (ID) |
| Metric | PPL |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 7 of 7 |
| 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

- [PPL](<https://www.sota2.com/research/sota/style-customization-on-20-users-opt-6-7b-based-id.md?metric=PPL>): Lower is better; unit: SCALAR.

- [Accuracy](<https://www.sota2.com/research/sota/style-customization-on-20-users-opt-6-7b-based-id.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | PPL | Result date | Report paper |
| --- | --- | --- | --- |
| X-Prompt (w/o CAL) | 29.7 | Not recorded | 68fa715c1e209131cea50dcb |
| Prompt tuning | 29.9 | Not recorded | 68fa715c1e209131cea50dcb |
| X-Prompt | 30.8 | Not recorded | 68fa715c1e209131cea50dcb |
| 32-shot | 34 | Not recorded | 68fa715c1e209131cea50dcb |
| 16-shot | 35.5 | Not recorded | 68fa715c1e209131cea50dcb |
| 8-shot | 36 | Not recorded | 68fa715c1e209131cea50dcb |
| No prompt | 38.9 | Not recorded | 68fa715c1e209131cea50dcb |

### X-Prompt (w/o CAL)

PPL: 29.7. Result ID: 6991ade33a6c31484909dbc5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715c1e209131cea50dcb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa715c1e209131cea50dcb.md>)

Recorded evaluation configuration: {"Backbone":"OPT-6.7b","Ablation":"without Content Augmentation for Language"}. This record does not assert independent reproduction.

### Prompt tuning

PPL: 29.9. Result ID: 6991ade33a6c31484909dbc3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715c1e209131cea50dcb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa690f1e209131cea1fe0e.md>)

Recorded evaluation configuration: {"Backbone":"OPT-6.7b"}. This record does not assert independent reproduction.

### X-Prompt

PPL: 30.8. Result ID: 6991ade33a6c31484909dbc4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715c1e209131cea50dcb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa715c1e209131cea50dcb.md>)

Recorded evaluation configuration: {"Backbone":"OPT-6.7b"}. This record does not assert independent reproduction.

### 32-shot

PPL: 34. Result ID: 6991ade33a6c31484909dbc2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715c1e209131cea50dcb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa680c1e209131cea19d53.md>)

[Implementation](<https://github.com/openai/gpt-3/blob/master/overlap_frequency.md>)

Recorded evaluation configuration: {"Backbone":"OPT-6.7b","k-shot":32}. This record does not assert independent reproduction.

### 16-shot

PPL: 35.5. Result ID: 6991ade33a6c31484909dbc1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715c1e209131cea50dcb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa680c1e209131cea19d53.md>)

[Implementation](<https://github.com/openai/gpt-3/blob/master/overlap_frequency.md>)

Recorded evaluation configuration: {"Backbone":"OPT-6.7b","k-shot":16}. This record does not assert independent reproduction.

### 8-shot

PPL: 36. Result ID: 6991ade33a6c31484909dbc0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715c1e209131cea50dcb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa680c1e209131cea19d53.md>)

[Implementation](<https://github.com/openai/gpt-3/blob/master/overlap_frequency.md>)

Recorded evaluation configuration: {"Backbone":"OPT-6.7b","k-shot":8}. This record does not assert independent reproduction.

### No prompt

PPL: 38.9. Result ID: 6991ade33a6c31484909dbbf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715c1e209131cea50dcb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cda1e209131cea36474.md>)

[Implementation](<https://github.com/facebookresearch/metaseq>)

Recorded evaluation configuration: {"Backbone":"OPT-6.7b","Prompt Type":"None"}. This record does not assert independent reproduction.

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