# Dialogue response evaluation on DSTC 12 (test)

Source: [Dialogue response evaluation on DSTC 12 (test)](<https://www.sota2.com/research/sota/dialogue-response-evaluation-on-dstc-12-test>)

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

Data updated: 2026-03-30T06:30:01.735Z

Evaluation of dialogue response quality on the test split of the DSTC 12 dataset, measured using specific human rating criteria such as Empathy, Trust, Skill, and Overall performance.

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 response evaluation |
| Dataset (including any stated split) | DSTC 12 (test) |
| Metric | Empathy |
| 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

- [Empathy](<https://www.sota2.com/research/sota/dialogue-response-evaluation-on-dstc-12-test.md?metric=Empathy>): Higher is better; unit: SCALAR.

- [Trust](<https://www.sota2.com/research/sota/dialogue-response-evaluation-on-dstc-12-test.md?metric=Trust>): Higher is better; unit: SCALAR.

- [Skill](<https://www.sota2.com/research/sota/dialogue-response-evaluation-on-dstc-12-test.md?metric=Skill>): Higher is better; unit: SCALAR.

- [Talent](<https://www.sota2.com/research/sota/dialogue-response-evaluation-on-dstc-12-test.md?metric=Talent>): Higher is better; unit: SCALAR.

- [Capability](<https://www.sota2.com/research/sota/dialogue-response-evaluation-on-dstc-12-test.md?metric=Capability>): Higher is better; unit: SCALAR.

- [Relevance](<https://www.sota2.com/research/sota/dialogue-response-evaluation-on-dstc-12-test.md?metric=Relevance>): Higher is better; unit: SCALAR.

- [NonRepetition](<https://www.sota2.com/research/sota/dialogue-response-evaluation-on-dstc-12-test.md?metric=NonRepetition>): Higher is better; unit: SCALAR.

- [Proactivity](<https://www.sota2.com/research/sota/dialogue-response-evaluation-on-dstc-12-test.md?metric=Proactivity>): Higher is better; unit: SCALAR.

- [Curiosity](<https://www.sota2.com/research/sota/dialogue-response-evaluation-on-dstc-12-test.md?metric=Curiosity>): Higher is better; unit: SCALAR.

- [Overall Score](<https://www.sota2.com/research/sota/dialogue-response-evaluation-on-dstc-12-test.md?metric=Overall%20Score>): Higher is better; unit: SCALAR.

- [Average Absolute Score](<https://www.sota2.com/research/sota/dialogue-response-evaluation-on-dstc-12-test.md?metric=Average%20Absolute%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Empathy | Result date | Report paper |
| --- | --- | --- | --- |
| Regression | 0.17 | Not recorded | 68fa7b0e1e209131cea866f7 |
| Baseline | 0.06 | Not recorded | 68fa7b0e1e209131cea866f7 |
| LM prompting | -0.08 | Not recorded | 68fa7b0e1e209131cea866f7 |
| Hybrid | -0.08 | Not recorded | 68fa7b0e1e209131cea866f7 |
| Classification | -0.17 | Not recorded | 68fa7b0e1e209131cea866f7 |

### Regression

Empathy: 0.17. Result ID: 69ca186347f34ce924b201cc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b0e1e209131cea866f7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b0e1e209131cea866f7.md>)

Recorded evaluation configuration: {"Method Approach":"Trained Regression Model"}. This record does not assert independent reproduction.

### Baseline

Empathy: 0.06. Result ID: 69ca186347f34ce924b201cf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b0e1e209131cea866f7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b6a1e209131cea884b0.md>)

Recorded evaluation configuration: {"Method Approach":"DSTC-12 Organizer Baseline"}. This record does not assert independent reproduction.

### LM prompting

Empathy: -0.08. Result ID: 69ca186347f34ce924b201cb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b0e1e209131cea866f7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b0e1e209131cea866f7.md>)

Recorded evaluation configuration: {"Method Approach":"Large Language Model Prompting"}. This record does not assert independent reproduction.

### Hybrid

Empathy: -0.08. Result ID: 69ca186347f34ce924b201ce.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b0e1e209131cea866f7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b0e1e209131cea866f7.md>)

Recorded evaluation configuration: {"Method Approach":"Combined Prompting and Regression"}. This record does not assert independent reproduction.

### Classification

Empathy: -0.17. Result ID: 69ca186347f34ce924b201cd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b0e1e209131cea866f7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b0e1e209131cea866f7.md>)

Recorded evaluation configuration: {"Method Approach":"Trained Classification Model"}. This record does not assert independent reproduction.

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