# Interruption on InstructS2S-Eval

Source: [Interruption on InstructS2S-Eval](<https://www.sota2.com/research/sota/interruption-on-instructs2s-eval>)

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

Data updated: 2026-06-15T07:53:54.948Z

Evaluation of model performance on handling interruptions using the InstructS2S-Eval dataset, measured by Overlap (s), ISR@2s, and Q2 S2S 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 | Interruption |
| Dataset (including any stated split) | InstructS2S-Eval |
| Metric | Overlap (s) |
| Unit | TIME |
| Direction | Lower is better |
| Results loaded | 3 of 3 |
| 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

- [Overlap (s)](<https://www.sota2.com/research/sota/interruption-on-instructs2s-eval.md?metric=Overlap%20(s)>): Lower is better; unit: TIME.

- [ISR@2s](<https://www.sota2.com/research/sota/interruption-on-instructs2s-eval.md?metric=ISR%402s>): Higher is better; unit: PERCENT.

- [Q2 S2S](<https://www.sota2.com/research/sota/interruption-on-instructs2s-eval.md?metric=Q2%20S2S>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Overlap (s) | Result date | Report paper |
| --- | --- | --- | --- |
| BayLing-Duplex | 1.1 | Not recorded | 6a2f8735d33dfb23e783c12c |
| BayLing-Duplex | 1.51 | Not recorded | 6a2f8735d33dfb23e783c12c |
| Moshi | 2.07 | Not recorded | 6a2f8735d33dfb23e783c12c |

### BayLing-Duplex

Overlap (s): 1.1. Result ID: 6a2faf77fbc373592c2bca6c.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8735d33dfb23e783c12c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2f8735d33dfb23e783c12c.md>)

Recorded evaluation configuration: {"Training Protocol":"SFT+DPO"}. This record does not assert independent reproduction.

### BayLing-Duplex

Overlap (s): 1.51. Result ID: 6a2faf77fbc373592c2bca6b.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8735d33dfb23e783c12c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2f8735d33dfb23e783c12c.md>)

Recorded evaluation configuration: {"Training Protocol":"SFT"}. This record does not assert independent reproduction.

### Moshi

Overlap (s): 2.07. Result ID: 6a2faf77fbc373592c2bca6a.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8735d33dfb23e783c12c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75001e209131cea652b7.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.
