# Persona-grounded Dialogue on FoCus (test)

Source: [Persona-grounded Dialogue on FoCus (test)](<https://www.sota2.com/research/sota/persona-grounded-dialogue-on-focus-test>)

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

Data updated: 2026-02-26T05:39:28.468Z

Evaluation of dialogue generation performance on the test split of the FoCus dataset, focusing on persona grounding and response quality using metrics such as Coh-UniEval, C Score, UE Score, and P Score.

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 | Persona-grounded Dialogue |
| Dataset (including any stated split) | FoCus (test) |
| Metric | Coh-UniEval |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [Coh-UniEval](<https://www.sota2.com/research/sota/persona-grounded-dialogue-on-focus-test.md?metric=Coh-UniEval>): Higher is better; unit: SCALAR.

- [C Score](<https://www.sota2.com/research/sota/persona-grounded-dialogue-on-focus-test.md?metric=C%20Score>): Higher is better; unit: SCALAR.

- [UE Score](<https://www.sota2.com/research/sota/persona-grounded-dialogue-on-focus-test.md?metric=UE%20Score>): Higher is better; unit: SCALAR.

- [P Score](<https://www.sota2.com/research/sota/persona-grounded-dialogue-on-focus-test.md?metric=P%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Coh-UniEval | Result date | Report paper |
| --- | --- | --- | --- |
| DPO | 0.98 | Not recorded | 69842421a8b01e76cc8ee7ae |
| PersoDPO | 0.7 | Not recorded | 69842421a8b01e76cc8ee7ae |
| Llama3.1 8B | 0.55 | Not recorded | 69842421a8b01e76cc8ee7ae |
| Mistral 7B | 0.53 | Not recorded | 69842421a8b01e76cc8ee7ae |
| Qwen2 7B | 0.37 | Not recorded | 69842421a8b01e76cc8ee7ae |

### DPO

Coh-UniEval: 0.98. Result ID: 69964cb75698665ec4912948.

[Reporting paper](<https://www.sota2.com/research/paper/69842421a8b01e76cc8ee7ae.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74511e209131cea6105d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2 5B"}. This record does not assert independent reproduction.

### PersoDPO

Coh-UniEval: 0.7. Result ID: 69964cb75698665ec4912949.

[Reporting paper](<https://www.sota2.com/research/paper/69842421a8b01e76cc8ee7ae.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69842421a8b01e76cc8ee7ae.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2 5B"}. This record does not assert independent reproduction.

### Llama3.1 8B

Coh-UniEval: 0.55. Result ID: 69964cb75698665ec4912947.

[Reporting paper](<https://www.sota2.com/research/paper/69842421a8b01e76cc8ee7ae.md>)

Recorded evaluation configuration: {"Mode":"Benchmark"}. This record does not assert independent reproduction.

### Mistral 7B

Coh-UniEval: 0.53. Result ID: 69964cb75698665ec4912946.

[Reporting paper](<https://www.sota2.com/research/paper/69842421a8b01e76cc8ee7ae.md>)

Recorded evaluation configuration: {"Mode":"Benchmark"}. This record does not assert independent reproduction.

### Qwen2 7B

Coh-UniEval: 0.37. Result ID: 69964cb75698665ec4912945.

[Reporting paper](<https://www.sota2.com/research/paper/69842421a8b01e76cc8ee7ae.md>)

Recorded evaluation configuration: {"Mode":"Benchmark"}. This record does not assert independent reproduction.

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