# Response Diversity on Role-playing dataset

Source: [Response Diversity on Role-playing dataset](<https://www.sota2.com/research/sota/response-diversity-on-role-playing-dataset>)

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

Data updated: 2026-05-19T12:42:56.846Z

Evaluation of model response diversity on the Role-playing dataset, measured using metrics such as Distinct-2, SNND, and NoC.

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 | Response Diversity |
| Dataset (including any stated split) | Role-playing dataset |
| Metric | Distinct-2 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 12 of 12 |
| 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

- [Distinct-2](<https://www.sota2.com/research/sota/response-diversity-on-role-playing-dataset.md?metric=Distinct-2>): Higher is better; unit: PERCENT.

- [SNND](<https://www.sota2.com/research/sota/response-diversity-on-role-playing-dataset.md?metric=SNND>): Lower is better; unit: PERCENT.

- [NoC](<https://www.sota2.com/research/sota/response-diversity-on-role-playing-dataset.md?metric=NoC>): Higher is better; unit: SCALAR.

- [Average Score](<https://www.sota2.com/research/sota/response-diversity-on-role-playing-dataset.md?metric=Average%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Distinct-2 | Result date | Report paper |
| --- | --- | --- | --- |
| w/o RL | 68 | Not recorded | 6a0bef08dbc836b87d82c9ae |
| w/o RL | 67 | Not recorded | 6a0bef08dbc836b87d82c9ae |
| GRPO | 64 | Not recorded | 6a0bef08dbc836b87d82c9ae |
| GRPO | 63 | Not recorded | 6a0bef08dbc836b87d82c9ae |
| PPO | 61 | Not recorded | 6a0bef08dbc836b87d82c9ae |
| PPR-GDE | 61 | Not recorded | 6a0bef08dbc836b87d82c9ae |
| w/o RL | 60 | Not recorded | 6a0bef08dbc836b87d82c9ae |
| PPO | 55 | Not recorded | 6a0bef08dbc836b87d82c9ae |
| PPR-GDE | 55 | Not recorded | 6a0bef08dbc836b87d82c9ae |
| PPR-GDE | 50 | Not recorded | 6a0bef08dbc836b87d82c9ae |
| PPO | 47 | Not recorded | 6a0bef08dbc836b87d82c9ae |
| GRPO | 20 | Not recorded | 6a0bef08dbc836b87d82c9ae |

### w/o RL

Distinct-2: 68. Result ID: 6a0c5aaefbc373592c03e1d5.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bef08dbc836b87d82c9ae.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-1.5B-Instruct"}. This record does not assert independent reproduction.

### w/o RL

Distinct-2: 67. Result ID: 6a0c5aaefbc373592c03e1d1.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bef08dbc836b87d82c9ae.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct"}. This record does not assert independent reproduction.

### GRPO

Distinct-2: 64. Result ID: 6a0c5aaefbc373592c03e1d3.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bef08dbc836b87d82c9ae.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

[Implementation](<https://github.com/deepseek-ai/DeepSeek-Math>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct"}. This record does not assert independent reproduction.

### GRPO

Distinct-2: 63. Result ID: 6a0c5aaefbc373592c03e1d7.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bef08dbc836b87d82c9ae.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

[Implementation](<https://github.com/deepseek-ai/DeepSeek-Math>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-1.5B-Instruct"}. This record does not assert independent reproduction.

### PPO

Distinct-2: 61. Result ID: 6a0c5aaefbc373592c03e1d2.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bef08dbc836b87d82c9ae.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63a01e209131ce9ffdaa.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct"}. This record does not assert independent reproduction.

### PPR-GDE

Distinct-2: 61. Result ID: 6a0c5aaefbc373592c03e1d4.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bef08dbc836b87d82c9ae.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0bef08dbc836b87d82c9ae.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct"}. This record does not assert independent reproduction.

### w/o RL

Distinct-2: 60. Result ID: 6a0c5aaefbc373592c03e1d9.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bef08dbc836b87d82c9ae.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-0.6B"}. This record does not assert independent reproduction.

### PPO

Distinct-2: 55. Result ID: 6a0c5aaefbc373592c03e1d6.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bef08dbc836b87d82c9ae.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63a01e209131ce9ffdaa.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-1.5B-Instruct"}. This record does not assert independent reproduction.

### PPR-GDE

Distinct-2: 55. Result ID: 6a0c5aaefbc373592c03e1d8.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bef08dbc836b87d82c9ae.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0bef08dbc836b87d82c9ae.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-1.5B-Instruct"}. This record does not assert independent reproduction.

### PPR-GDE

Distinct-2: 50. Result ID: 6a0c5aaefbc373592c03e1dc.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bef08dbc836b87d82c9ae.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0bef08dbc836b87d82c9ae.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-0.6B"}. This record does not assert independent reproduction.

### PPO

Distinct-2: 47. Result ID: 6a0c5aaefbc373592c03e1da.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bef08dbc836b87d82c9ae.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63a01e209131ce9ffdaa.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-0.6B"}. This record does not assert independent reproduction.

### GRPO

Distinct-2: 20. Result ID: 6a0c5aaefbc373592c03e1db.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bef08dbc836b87d82c9ae.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

[Implementation](<https://github.com/deepseek-ai/DeepSeek-Math>)

Recorded evaluation configuration: {"Backbone":"Qwen3-0.6B"}. This record does not assert independent reproduction.

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