# Conversational Derailment Prediction on CGA-CMV-Large (test)

Source: [Conversational Derailment Prediction on CGA-CMV-Large (test)](<https://www.sota2.com/research/sota/conversational-derailment-prediction-on-cga-cmv-large-test>)

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

Data updated: 2026-05-29T08:23:00.375Z

Evaluation of model performance on predicting conversational derailment using the test split of the CGA-CMV-Large dataset, assessed using accuracy, precision, recall, F1 score, and related 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 | Conversational Derailment Prediction |
| Dataset (including any stated split) | CGA-CMV-Large (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/conversational-derailment-prediction-on-cga-cmv-large-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [False Positive Rate](<https://www.sota2.com/research/sota/conversational-derailment-prediction-on-cga-cmv-large-test.md?metric=False%20Positive%20Rate>): Lower is better; unit: PERCENT.

- [Precision](<https://www.sota2.com/research/sota/conversational-derailment-prediction-on-cga-cmv-large-test.md?metric=Precision>): Higher is better; unit: PERCENT.

- [Recall](<https://www.sota2.com/research/sota/conversational-derailment-prediction-on-cga-cmv-large-test.md?metric=Recall>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/conversational-derailment-prediction-on-cga-cmv-large-test.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [H Score](<https://www.sota2.com/research/sota/conversational-derailment-prediction-on-cga-cmv-large-test.md?metric=H%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| SOTA | 70.9 | Not recorded | 6a191dabdbd7150506f828f2 |
| SOTA + selective deferral | 70.9 | Not recorded | 6a191dabdbd7150506f828f2 |
| SOTA + simulation (average) | 70.2 | Not recorded | 6a191dabdbd7150506f828f2 |
| SOTA + simulation (majority) | 70 | Not recorded | 6a191dabdbd7150506f828f2 |
| SOTA (oracle threshold) | 70 | Not recorded | 6a191dabdbd7150506f828f2 |
| SOTA + random deferral | 69.4 | Not recorded | 6a191dabdbd7150506f828f2 |

### SOTA

Accuracy: 70.9. Result ID: 6a194b6bfbc373592c1391b0.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dabdbd7150506f828f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75001e209131cea6527c.md>)

Recorded evaluation configuration: {"Decision Mechanism":"Standard"}. This record does not assert independent reproduction.

### SOTA + selective deferral

Accuracy: 70.9. Result ID: 6a194b6bfbc373592c1391b1.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dabdbd7150506f828f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a191dabdbd7150506f828f2.md>)

Recorded evaluation configuration: {"Decision Mechanism":"Selective Deferral"}. This record does not assert independent reproduction.

### SOTA + simulation (average)

Accuracy: 70.2. Result ID: 6a194b6bfbc373592c1391b3.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dabdbd7150506f828f2.md>)

Recorded evaluation configuration: {"Decision Mechanism":"Simulation (Average)"}. This record does not assert independent reproduction.

### SOTA + simulation (majority)

Accuracy: 70. Result ID: 6a194b6bfbc373592c1391b4.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dabdbd7150506f828f2.md>)

Recorded evaluation configuration: {"Decision Mechanism":"Simulation (Majority)"}. This record does not assert independent reproduction.

### SOTA (oracle threshold)

Accuracy: 70. Result ID: 6a194b6bfbc373592c1391b5.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dabdbd7150506f828f2.md>)

Recorded evaluation configuration: {"Decision Mechanism":"Oracle Threshold Selection"}. This record does not assert independent reproduction.

### SOTA + random deferral

Accuracy: 69.4. Result ID: 6a194b6bfbc373592c1391b2.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dabdbd7150506f828f2.md>)

Recorded evaluation configuration: {"Decision Mechanism":"Random Deferral"}. This record does not assert independent reproduction.

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