# Dialogue Disentanglement on Ubuntu IRC (test)

Source: [Dialogue Disentanglement on Ubuntu IRC (test)](<https://www.sota2.com/research/sota/dialogue-disentanglement-on-ubuntu-irc-test>)

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

Data updated: 2026-02-26T06:08:01.742Z

Evaluation of Dialogue Disentanglement performance on the test split of the Ubuntu IRC dataset, measured using metrics including VI, ARI, 1-1 Match Rate, Precision, Recall, and F1 Score.

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 Disentanglement |
| Dataset (including any stated split) | Ubuntu IRC (test) |
| Metric | VI |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 17 of 17 |
| 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

- [VI](<https://www.sota2.com/research/sota/dialogue-disentanglement-on-ubuntu-irc-test.md?metric=VI>): Higher is better; unit: PERCENT.

- [ARI](<https://www.sota2.com/research/sota/dialogue-disentanglement-on-ubuntu-irc-test.md?metric=ARI>): Higher is better; unit: SCALAR.

- [1-1 Match Rate](<https://www.sota2.com/research/sota/dialogue-disentanglement-on-ubuntu-irc-test.md?metric=1-1%20Match%20Rate>): Higher is better; unit: PERCENT.

- [F1](<https://www.sota2.com/research/sota/dialogue-disentanglement-on-ubuntu-irc-test.md?metric=F1>): Higher is better; unit: PERCENT.

- [Precision](<https://www.sota2.com/research/sota/dialogue-disentanglement-on-ubuntu-irc-test.md?metric=Precision>): Higher is better; unit: PERCENT.

- [Recall](<https://www.sota2.com/research/sota/dialogue-disentanglement-on-ubuntu-irc-test.md?metric=Recall>): Higher is better; unit: PERCENT.

## Recorded results

| Method | VI | Result date | Report paper |
| --- | --- | --- | --- |
| Structural Characterization model | 94.6 | Not recorded | 68fa6c011e209131cea319d2 |
| Ptr-Net | 94.2 | Not recorded | 68fa6c011e209131cea319d2 |
| DialBERT | 93.2 | Not recorded | 68fa6c011e209131cea319d2 |
| Ptr-Net | 93.1 | Not recorded | 68fa6c011e209131cea319d2 |
| Ptr-Net | 93 | Not recorded | 68fa6c011e209131cea319d2 |
| DialBERT | 92.6 | Not recorded | 68fa6c011e209131cea319d2 |
| DialBERT | 92.4 | Not recorded | 68fa6c011e209131cea319d2 |
| DialBERT | 92.3 | Not recorded | 68fa6c011e209131cea319d2 |
| Ptr-Net | 92.3 | Not recorded | 68fa6c011e209131cea319d2 |
| FeedForward | 91.5 | Not recorded | 68fa6c011e209131cea319d2 |
| BERT\_base | 91.4 | Not recorded | 68fa6c011e209131cea319d2 |
| FeedForward | 91.3 | Not recorded | 68fa6c011e209131cea319d2 |
| BERT | 90.8 | Not recorded | 68fa6c011e209131cea319d2 |
| FeedForward | 86.2 | Not recorded | 68fa6c011e209131cea319d2 |
| Elsner | 82.1 | Not recorded | 68fa6c011e209131cea319d2 |
| Lowe | 80.6 | Not recorded | 68fa6c011e209131cea319d2 |
| FeedForward | 69.3 | Not recorded | 68fa6c011e209131cea319d2 |

### Structural Characterization model

VI: 94.6. Result ID: 698eba10ffc3901de0a02280.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

[Implementation](<https://github.com/xbmxb/StructureCharacterization4DD>)

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

### Ptr-Net

VI: 94.2. Result ID: 698eba10ffc3901de0a0227e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68891e209131cea1cc00.md>)

Recorded evaluation configuration: {"components":"Joint train & Self-link"}. This record does not assert independent reproduction.

### DialBERT

VI: 93.2. Result ID: 698eba10ffc3901de0a02278.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ab21e209131cea29d4d.md>)

Recorded evaluation configuration: {"features":"cov"}. This record does not assert independent reproduction.

### Ptr-Net

VI: 93.1. Result ID: 698eba10ffc3901de0a0227c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68891e209131cea1cc00.md>)

Recorded evaluation configuration: {"training":"Joint train"}. This record does not assert independent reproduction.

### Ptr-Net

VI: 93. Result ID: 698eba10ffc3901de0a0227d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68891e209131cea1cc00.md>)

Recorded evaluation configuration: {"components":"Self-link"}. This record does not assert independent reproduction.

### DialBERT

VI: 92.6. Result ID: 698eba10ffc3901de0a02277.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ab21e209131cea29d4d.md>)

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

### DialBERT

VI: 92.4. Result ID: 698eba10ffc3901de0a02279.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ab21e209131cea29d4d.md>)

Recorded evaluation configuration: {"features":"feature"}. This record does not assert independent reproduction.

### DialBERT

VI: 92.3. Result ID: 698eba10ffc3901de0a0227a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ab21e209131cea29d4d.md>)

Recorded evaluation configuration: {"features":"future context"}. This record does not assert independent reproduction.

### Ptr-Net

VI: 92.3. Result ID: 698eba10ffc3901de0a0227b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68891e209131cea1cc00.md>)

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

### FeedForward

VI: 91.5. Result ID: 698eba10ffc3901de0a02272.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67a71e209131cea177d9.md>)

[Implementation](<https://jkk.name/irc-disentanglement>)

Recorded evaluation configuration: {"aggregation":"10 vote"}. This record does not assert independent reproduction.

### BERT\_base

VI: 91.4. Result ID: 698eba10ffc3901de0a0227f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c31e209131cea0c4ed.md>)

[Implementation](<https://github.com/google-research/bert>)

Recorded evaluation configuration: {"note":"Our baseline"}. This record does not assert independent reproduction.

### FeedForward

VI: 91.3. Result ID: 698eba10ffc3901de0a02270.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67a71e209131cea177d9.md>)

[Implementation](<https://jkk.name/irc-disentanglement>)

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

### BERT

VI: 90.8. Result ID: 698eba10ffc3901de0a02276.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ab21e209131cea29d4d.md>)

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

### FeedForward

VI: 86.2. Result ID: 698eba10ffc3901de0a02271.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67a71e209131cea177d9.md>)

[Implementation](<https://jkk.name/irc-disentanglement>)

Recorded evaluation configuration: {"aggregation":"10 union"}. This record does not assert independent reproduction.

### Elsner

VI: 82.1. Result ID: 698eba10ffc3901de0a02274.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

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

### Lowe

VI: 80.6. Result ID: 698eba10ffc3901de0a02275.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

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

### FeedForward

VI: 69.3. Result ID: 698eba10ffc3901de0a02273.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67a71e209131cea177d9.md>)

[Implementation](<https://jkk.name/irc-disentanglement>)

Recorded evaluation configuration: {"aggregation":"10 intersect"}. This record does not assert independent reproduction.

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