# Linguistic Trait Steering on Open-ended Trait Steering Dataset

Source: [Linguistic Trait Steering on Open-ended Trait Steering Dataset](<https://www.sota2.com/research/sota/linguistic-trait-steering-on-open-ended-trait-steering-dataset>)

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

Data updated: 2026-03-11T08:30:11.276Z

Evaluation of model performance on steering specific linguistic traits, including Humor, Rudeness, Excitement, and Sadness, using the Open-ended Trait Steering Dataset.

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 | Linguistic Trait Steering |
| Dataset (including any stated split) | Open-ended Trait Steering Dataset |
| Metric | Humor Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Humor Score](<https://www.sota2.com/research/sota/linguistic-trait-steering-on-open-ended-trait-steering-dataset.md?metric=Humor%20Score>): Higher is better; unit: PERCENT.

- [Rudeness Score](<https://www.sota2.com/research/sota/linguistic-trait-steering-on-open-ended-trait-steering-dataset.md?metric=Rudeness%20Score>): Higher is better; unit: PERCENT.

- [Excitement Score](<https://www.sota2.com/research/sota/linguistic-trait-steering-on-open-ended-trait-steering-dataset.md?metric=Excitement%20Score>): Higher is better; unit: PERCENT.

- [Sadness Score](<https://www.sota2.com/research/sota/linguistic-trait-steering-on-open-ended-trait-steering-dataset.md?metric=Sadness%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Humor Score | Result date | Report paper |
| --- | --- | --- | --- |
| Linear | 85 | Not recorded | 69b0f7252d603e6de3e45ec6 |
| Curveball | 75 | Not recorded | 69b0f7252d603e6de3e45ec6 |
| Linear | 54.9 | Not recorded | 69b0f7252d603e6de3e45ec6 |
| Curveball | 28.2 | Not recorded | 69b0f7252d603e6de3e45ec6 |

### Linear

Humor Score: 85. Result ID: 69b1280f6dbb7075b95b8994.

[Reporting paper](<https://www.sota2.com/research/paper/69b0f7252d603e6de3e45ec6.md>)

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

Recorded evaluation configuration: {"Model":"Phi-3.5-mini-It"}. This record does not assert independent reproduction.

### Curveball

Humor Score: 75. Result ID: 69b1280f6dbb7075b95b8995.

[Reporting paper](<https://www.sota2.com/research/paper/69b0f7252d603e6de3e45ec6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b0f7252d603e6de3e45ec6.md>)

Recorded evaluation configuration: {"Model":"Phi-3.5-mini-It"}. This record does not assert independent reproduction.

### Linear

Humor Score: 54.9. Result ID: 69b1280f6dbb7075b95b8992.

[Reporting paper](<https://www.sota2.com/research/paper/69b0f7252d603e6de3e45ec6.md>)

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

Recorded evaluation configuration: {"Model":"Llama-3.2-1B-It"}. This record does not assert independent reproduction.

### Curveball

Humor Score: 28.2. Result ID: 69b1280f6dbb7075b95b8993.

[Reporting paper](<https://www.sota2.com/research/paper/69b0f7252d603e6de3e45ec6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b0f7252d603e6de3e45ec6.md>)

Recorded evaluation configuration: {"Model":"Llama-3.2-1B-It"}. This record does not assert independent reproduction.

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