# Dialogue Generation on Movie Dic (test)

Source: [Dialogue Generation on Movie Dic (test)](<https://www.sota2.com/research/sota/dialogue-generation-on-movie-dic-test>)

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

Data updated: 2026-02-26T06:00:41.908Z

Evaluation of Dialogue Generation performance on the test split of the Movie Dic dataset, measured using ROUGE-L and Perplexity (PPL).

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 | Dialogue Generation |
| Dataset (including any stated split) | Movie Dic (test) |
| Metric | ROUGE-L |
| Unit | PERCENT |
| 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

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

- [Perplexity](<https://www.sota2.com/research/sota/dialogue-generation-on-movie-dic-test.md?metric=Perplexity>): Lower is better; unit: SCALAR.

## Recorded results

| Method | ROUGE-L | Result date | Report paper |
| --- | --- | --- | --- |
| SECOND THOUGHTS (AEM + VM default) | 17.35 | Not recorded | 68fa6e5a1e209131cea3f4b9 |
| SECOND THOUGHTS (AEM + AIL) | 15.02 | Not recorded | 68fa6e5a1e209131cea3f4b9 |
| SECOND THOUGHTS (AEM Only) | 14 | Not recorded | 68fa6e5a1e209131cea3f4b9 |
| InstructGPT | 11.47 | Not recorded | 68fa6e5a1e209131cea3f4b9 |
| GPT-3 | 10.26 | Not recorded | 68fa6e5a1e209131cea3f4b9 |

### SECOND THOUGHTS (AEM + VM default)

ROUGE-L: 17.35. Result ID: 69902c463a6c314849fea2f2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4b9.md>)

Recorded evaluation configuration: {"variant":"AEM + VM (default)"}. This record does not assert independent reproduction.

### SECOND THOUGHTS (AEM + AIL)

ROUGE-L: 15.02. Result ID: 69902c463a6c314849fea2f3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4b9.md>)

Recorded evaluation configuration: {"variant":"AEM + AIL"}. This record does not assert independent reproduction.

### SECOND THOUGHTS (AEM Only)

ROUGE-L: 14. Result ID: 69902c463a6c314849fea2f4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4b9.md>)

Recorded evaluation configuration: {"variant":"AEM Only"}. This record does not assert independent reproduction.

### InstructGPT

ROUGE-L: 11.47. Result ID: 69902c463a6c314849fea2f6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bf71e209131cea314c0.md>)

[Implementation](<https://github.com/openai/following-instructions-human-feedback>)

Recorded evaluation configuration: {"Service":"Huge LM API service"}. This record does not assert independent reproduction.

### GPT-3

ROUGE-L: 10.26. Result ID: 69902c463a6c314849fea2f5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4b9.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: {"Service":"Huge LM API service"}. This record does not assert independent reproduction.

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