# Interruption Handling on Full-Duplex Bench v1.5

Source: [Interruption Handling on Full-Duplex Bench v1.5](<https://www.sota2.com/research/sota/interruption-handling-on-full-duplex-bench-v1-5>)

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

Data updated: 2026-05-12T15:02:43.592Z

Evaluation of model performance on handling user interruptions within the Full-Duplex Bench v1.5 dataset, focusing on response success rates (Respond, Resume, Uncertain, Unknown) and associated latencies.

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) | Full-Duplex Bench v1.5 |
| Metric | Respond Rate |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Respond Rate](<https://www.sota2.com/research/sota/interruption-handling-on-full-duplex-bench-v1-5.md?metric=Respond%20Rate>): Higher is better; unit: PERCENT.

- [Resume Rate](<https://www.sota2.com/research/sota/interruption-handling-on-full-duplex-bench-v1-5.md?metric=Resume%20Rate>): Higher is better; unit: PERCENT.

- [Uncertain Rate](<https://www.sota2.com/research/sota/interruption-handling-on-full-duplex-bench-v1-5.md?metric=Uncertain%20Rate>): Lower is better; unit: PERCENT.

- [Unknown Rate](<https://www.sota2.com/research/sota/interruption-handling-on-full-duplex-bench-v1-5.md?metric=Unknown%20Rate>): Lower is better; unit: PERCENT.

- [Stop Latency (s)](<https://www.sota2.com/research/sota/interruption-handling-on-full-duplex-bench-v1-5.md?metric=Stop%20Latency%20(s)>): Lower is better; unit: TIME.

- [Respond Latency (s)](<https://www.sota2.com/research/sota/interruption-handling-on-full-duplex-bench-v1-5.md?metric=Respond%20Latency%20(s)>): Lower is better; unit: TIME.

## Recorded results

| Method | Respond Rate | Result date | Report paper |
| --- | --- | --- | --- |
| Freeze-Omni | 72 | Not recorded | 6a02b49bdbc836b87d824e14 |
| CF-Duplex | 72 | Not recorded | 6a02b49bdbc836b87d824e14 |
| Moshi | 50 | Not recorded | 6a02b49bdbc836b87d824e14 |
| XA-Duplex | 32 | Not recorded | 6a02b49bdbc836b87d824e14 |

### Freeze-Omni

Respond Rate: 72. Result ID: 6a03407b47f34ce924e83ce1.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b49bdbc836b87d824e14.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa760a1e209131cea6b0d4.md>)

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

### CF-Duplex

Respond Rate: 72. Result ID: 6a03407b47f34ce924e83ce3.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b49bdbc836b87d824e14.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b49bdbc836b87d824e14.md>)

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

### Moshi

Respond Rate: 50. Result ID: 6a03407b47f34ce924e83ce2.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b49bdbc836b87d824e14.md>)

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

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

### XA-Duplex

Respond Rate: 32. Result ID: 6a03407b47f34ce924e83ce4.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b49bdbc836b87d824e14.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b49bdbc836b87d824e14.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.