# Dialogue Generation on PersonaChat

Source: [Dialogue Generation on PersonaChat](<https://www.sota2.com/research/sota/dialogue-generation-on-personachat>)

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

Data updated: 2026-02-26T05:37:12.855Z

Evaluation of dialogue response generation quality on the PersonaChat dataset, measured using BLEU scores, Distinctness metrics, and BLEURT.

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) | PersonaChat |
| Metric | BLEU-1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [BLEU-1](<https://www.sota2.com/research/sota/dialogue-generation-on-personachat.md?metric=BLEU-1>): Higher is better; unit: PERCENT.

- [BLEU-2](<https://www.sota2.com/research/sota/dialogue-generation-on-personachat.md?metric=BLEU-2>): Higher is better; unit: PERCENT.

- [BLEU-3](<https://www.sota2.com/research/sota/dialogue-generation-on-personachat.md?metric=BLEU-3>): Higher is better; unit: PERCENT.

- [BLEU-4](<https://www.sota2.com/research/sota/dialogue-generation-on-personachat.md?metric=BLEU-4>): Higher is better; unit: PERCENT.

- [DIST-1](<https://www.sota2.com/research/sota/dialogue-generation-on-personachat.md?metric=DIST-1>): Higher is better; unit: SCALAR.

- [DIST-2](<https://www.sota2.com/research/sota/dialogue-generation-on-personachat.md?metric=DIST-2>): Higher is better; unit: SCALAR.

- [BLEURT](<https://www.sota2.com/research/sota/dialogue-generation-on-personachat.md?metric=BLEURT>): Higher is better; unit: SCALAR.

## Recorded results

| Method | BLEU-1 | Result date | Report paper |
| --- | --- | --- | --- |
| DialoGPS | 19.05 | Not recorded | 68fa6feb1e209131cea48261 |
| DialoGPS | 19.04 | Not recorded | 68fa6feb1e209131cea48261 |
| DialoGPS | 18.96 | Not recorded | 68fa6feb1e209131cea48261 |
| M&D-D | 18.42 | Not recorded | 68fa6feb1e209131cea48261 |
| DialoGPS | 18.29 | Not recorded | 68fa6feb1e209131cea48261 |
| ResBag | 17.82 | Not recorded | 68fa6feb1e209131cea48261 |
| Transformer | 17.79 | Not recorded | 68fa6feb1e209131cea48261 |
| TSA | 17.76 | Not recorded | 68fa6feb1e209131cea48261 |

### DialoGPS

BLEU-1: 19.05. Result ID: 699102523a6c31484904fcba.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6feb1e209131cea48261.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6feb1e209131cea48261.md>)

Recorded evaluation configuration: {"K (Neighbor size)":4,"Decoding scheme":"Top-5 Sampling"}. This record does not assert independent reproduction.

### DialoGPS

BLEU-1: 19.04. Result ID: 699102523a6c31484904fcbb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6feb1e209131cea48261.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6feb1e209131cea48261.md>)

Recorded evaluation configuration: {"K (Neighbor size)":8,"Decoding scheme":"Top-5 Sampling"}. This record does not assert independent reproduction.

### DialoGPS

BLEU-1: 18.96. Result ID: 699102523a6c31484904fcb9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6feb1e209131cea48261.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6feb1e209131cea48261.md>)

Recorded evaluation configuration: {"K (Neighbor size)":2,"Decoding scheme":"Top-5 Sampling"}. This record does not assert independent reproduction.

### M&D-D

BLEU-1: 18.42. Result ID: 699102523a6c31484904fcb7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6feb1e209131cea48261.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68be1e209131cea1dfa3.md>)

Recorded evaluation configuration: {"Decoding scheme":"Top-5 Sampling"}. This record does not assert independent reproduction.

### DialoGPS

BLEU-1: 18.29. Result ID: 699102523a6c31484904fcb8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6feb1e209131cea48261.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6feb1e209131cea48261.md>)

Recorded evaluation configuration: {"K (Neighbor size)":1,"Decoding scheme":"Top-5 Sampling"}. This record does not assert independent reproduction.

### ResBag

BLEU-1: 17.82. Result ID: 699102523a6c31484904fcb5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6feb1e209131cea48261.md>)

Recorded evaluation configuration: {"Decoding scheme":"Top-5 Sampling"}. This record does not assert independent reproduction.

### Transformer

BLEU-1: 17.79. Result ID: 699102523a6c31484904fcb4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6feb1e209131cea48261.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa703d1e209131cea49db1.md>)

[Implementation](<https://github.com/tensorflow/tensor2tensor>)

Recorded evaluation configuration: {"Decoding scheme":"Top-5 Sampling"}. This record does not assert independent reproduction.

### TSA

BLEU-1: 17.76. Result ID: 699102523a6c31484904fcb6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6feb1e209131cea48261.md>)

Recorded evaluation configuration: {"Decoding scheme":"Top-5 Sampling"}. This record does not assert independent reproduction.

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