# Open-ended Dialogue Generation on CMD (test)

Source: [Open-ended Dialogue Generation on CMD (test)](<https://www.sota2.com/research/sota/open-ended-dialogue-generation-on-cmd-test>)

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

Data updated: 2026-05-28T08:11:42.953Z

Evaluation of open-ended dialogue generation performance on the test split of the CMD dataset, utilizing metrics such as Rouge-L, Reranker scores, and RL/RR metrics.

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 | Open-ended Dialogue Generation |
| Dataset (including any stated split) | CMD (test) |
| Metric | Rouge-L |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 26 of 26 |
| 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

- [Rouge-L](<https://www.sota2.com/research/sota/open-ended-dialogue-generation-on-cmd-test.md?metric=Rouge-L>): Higher is better; unit: PERCENT.

- [Reranker Score](<https://www.sota2.com/research/sota/open-ended-dialogue-generation-on-cmd-test.md?metric=Reranker%20Score>): Higher is better; unit: PERCENT.

- [RL@8](<https://www.sota2.com/research/sota/open-ended-dialogue-generation-on-cmd-test.md?metric=RL%408>): Higher is better; unit: PERCENT.

- [RR@8](<https://www.sota2.com/research/sota/open-ended-dialogue-generation-on-cmd-test.md?metric=RR%408>): Higher is better; unit: PERCENT.

- [Average Score](<https://www.sota2.com/research/sota/open-ended-dialogue-generation-on-cmd-test.md?metric=Average%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Rouge-L | Result date | Report paper |
| --- | --- | --- | --- |
| NSR | 19.4 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 19.4 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| NSR | 19.4 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 19.4 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 18.7 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| EAPO | 18.7 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 18.7 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| EAPO | 18.7 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| GRPO | 18.6 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 18.6 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| GRPO | 18.6 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 18.6 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 18.5 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 18.5 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 18.4 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 18.4 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 17.3 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 17.3 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 17 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 17 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 16 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 16 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 15.8 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| W-REINFORCE | 15.8 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| PSR | 14.7 | Not recorded | 6a17cc2bdbd7150506f813f2 |
| PSR | 14.7 | Not recorded | 6a17cc2bdbd7150506f813f2 |

### NSR

Rouge-L: 19.4. Result ID: 6a17f7d3fbc373592c119127.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 19.4. Result ID: 6a17f7d3fbc373592c119128.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"positive sample weight (w+)":"0.05","negative sample weight (w-)":"1"}. This record does not assert independent reproduction.

### NSR

Rouge-L: 19.4. Result ID: 6a17f7d3fbc373592c119141.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 19.4. Result ID: 6a17f7d3fbc373592c119142.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"w+":"0.05","w-":"1"}. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 18.7. Result ID: 6a17f7d3fbc373592c119130.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"positive sample weight (w+)":"1","negative sample weight (w-)":"5"}. This record does not assert independent reproduction.

### EAPO

Rouge-L: 18.7. Result ID: 6a17f7d3fbc373592c119131.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 18.7. Result ID: 6a17f7d3fbc373592c11914a.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"w+":"1","w-":"5"}. This record does not assert independent reproduction.

### EAPO

Rouge-L: 18.7. Result ID: 6a17f7d3fbc373592c11914b.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### GRPO

Rouge-L: 18.6. Result ID: 6a17f7d3fbc373592c119125.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76a91e209131cea6e3eb.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 18.6. Result ID: 6a17f7d3fbc373592c11912a.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"positive sample weight (w+)":"0.2","negative sample weight (w-)":"1"}. This record does not assert independent reproduction.

### GRPO

Rouge-L: 18.6. Result ID: 6a17f7d3fbc373592c11913f.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76a91e209131cea6e3eb.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 18.6. Result ID: 6a17f7d3fbc373592c119144.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"w+":"0.2","w-":"1"}. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 18.5. Result ID: 6a17f7d3fbc373592c119129.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"positive sample weight (w+)":"0.1","negative sample weight (w-)":"1"}. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 18.5. Result ID: 6a17f7d3fbc373592c119143.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"w+":"0.1","w-":"1"}. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 18.4. Result ID: 6a17f7d3fbc373592c11912b.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"positive sample weight (w+)":"1","negative sample weight (w-)":"1"}. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 18.4. Result ID: 6a17f7d3fbc373592c119145.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"w+":"1","w-":"1"}. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 17.3. Result ID: 6a17f7d3fbc373592c11912c.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"positive sample weight (w+)":"5","negative sample weight (w-)":"1"}. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 17.3. Result ID: 6a17f7d3fbc373592c119146.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"w+":"5","w-":"1"}. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 17. Result ID: 6a17f7d3fbc373592c11912f.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"positive sample weight (w+)":"1","negative sample weight (w-)":"0.2"}. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 17. Result ID: 6a17f7d3fbc373592c119149.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"w+":"1","w-":"0.2"}. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 16. Result ID: 6a17f7d3fbc373592c11912d.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"positive sample weight (w+)":"1","negative sample weight (w-)":"0.05"}. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 16. Result ID: 6a17f7d3fbc373592c119147.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"w+":"1","w-":"0.05"}. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 15.8. Result ID: 6a17f7d3fbc373592c11912e.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"positive sample weight (w+)":"1","negative sample weight (w-)":"0.1"}. This record does not assert independent reproduction.

### W-REINFORCE

Rouge-L: 15.8. Result ID: 6a17f7d3fbc373592c119148.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"w+":"1","w-":"0.1"}. This record does not assert independent reproduction.

### PSR

Rouge-L: 14.7. Result ID: 6a17f7d3fbc373592c119126.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### PSR

Rouge-L: 14.7. Result ID: 6a17f7d3fbc373592c119140.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc2bdbd7150506f813f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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