# Personality Steering on Big Five generation

Source: [Personality Steering on Big Five generation](<https://www.sota2.com/research/sota/personality-steering-on-big-five-generation>)

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

Data updated: 2026-05-08T10:28:08.884Z

Evaluation of model performance on the Personality Steering task using the Big Five generation dataset, measured by the GAS metric.

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 | Personality Steering |
| Dataset (including any stated split) | Big Five generation |
| Metric | GAS 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

- [GAS Score](<https://www.sota2.com/research/sota/personality-steering-on-big-five-generation.md?metric=GAS%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | GAS Score | Result date | Report paper |
| --- | --- | --- | --- |
| Memory Inception | 0.44 | Not recorded | 69fd6e1c5497b3b670c89492 |
| Prompt | 0.26 | Not recorded | 69fd6e1c5497b3b670c89492 |
| Prompt | -0.524 | Not recorded | 69fd6e1c5497b3b670c89492 |
| Memory Inception | -0.602 | Not recorded | 69fd6e1c5497b3b670c89492 |
| CAA | -0.874 | Not recorded | 69fd6e1c5497b3b670c89492 |
| CAA | -0.926 | Not recorded | 69fd6e1c5497b3b670c89492 |

### Memory Inception

GAS Score: 0.44. Result ID: 69fdbab147f34ce924e153a9.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e1c5497b3b670c89492.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e1c5497b3b670c89492.md>)

Recorded evaluation configuration: {"Model":"Qwen3-30B-A3B","Setting":"–"}. This record does not assert independent reproduction.

### Prompt

GAS Score: 0.26. Result ID: 69fdbab147f34ce924e153a7.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e1c5497b3b670c89492.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: {"Model":"Qwen3-30B-A3B","Setting":"–"}. This record does not assert independent reproduction.

### Prompt

GAS Score: -0.524. Result ID: 69fdbab147f34ce924e153a4.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e1c5497b3b670c89492.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: {"Model":"Llama3.1-8B","Setting":"–"}. This record does not assert independent reproduction.

### Memory Inception

GAS Score: -0.602. Result ID: 69fdbab147f34ce924e153a6.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e1c5497b3b670c89492.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e1c5497b3b670c89492.md>)

Recorded evaluation configuration: {"Model":"Llama3.1-8B","Setting":"–"}. This record does not assert independent reproduction.

### CAA

GAS Score: -0.874. Result ID: 69fdbab147f34ce924e153a5.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e1c5497b3b670c89492.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74051e209131cea5f4ad.md>)

Recorded evaluation configuration: {"Model":"Llama3.1-8B","Setting":"–"}. This record does not assert independent reproduction.

### CAA

GAS Score: -0.926. Result ID: 69fdbab147f34ce924e153a8.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e1c5497b3b670c89492.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74051e209131cea5f4ad.md>)

Recorded evaluation configuration: {"Model":"Qwen3-30B-A3B","Setting":"–"}. This record does not assert independent reproduction.

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