# Hallucination Detection on CWQ (F1, Precision, Recall, Accuracy, AUC-PR)

Source: [Hallucination Detection on CWQ (F1, Precision, Recall, Accuracy, AUC-PR)](<https://www.sota2.com/research/sota/hallucination-detection-on-cwq-f1-precision-recall-accuracy-auc-pr>)

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

Data updated: 2026-06-02T07:57:21.068Z

Evaluation of model performance on hallucination detection using the CWQ dataset, reporting F1, Precision, Recall, Accuracy, and AUC-PR.

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 | Hallucination Detection |
| Dataset (including any stated split) | CWQ |
| Metric | F1 Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 11 of 11 |
| 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

- [F1 Score](<https://www.sota2.com/research/sota/hallucination-detection-on-cwq-f1-precision-recall-accuracy-auc-pr.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [Precision](<https://www.sota2.com/research/sota/hallucination-detection-on-cwq-f1-precision-recall-accuracy-auc-pr.md?metric=Precision>): Higher is better; unit: PERCENT.

- [Recall](<https://www.sota2.com/research/sota/hallucination-detection-on-cwq-f1-precision-recall-accuracy-auc-pr.md?metric=Recall>): Higher is better; unit: PERCENT.

- [Accuracy](<https://www.sota2.com/research/sota/hallucination-detection-on-cwq-f1-precision-recall-accuracy-auc-pr.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [AUC-PR](<https://www.sota2.com/research/sota/hallucination-detection-on-cwq-f1-precision-recall-accuracy-auc-pr.md?metric=AUC-PR>): Higher is better; unit: PERCENT.

## Recorded results

| Method | F1 Score | Result date | Report paper |
| --- | --- | --- | --- |
| KG-Guard | 87.4 | Not recorded | 6a1e63b3dbd7150506f85292 |
| KG-Guard | 86.3 | Not recorded | 6a1e63b3dbd7150506f85292 |
| GPT 5 Mini | 84.2 | Not recorded | 6a1e63b3dbd7150506f85292 |
| Most Frequent | 83.5 | Not recorded | 6a1e63b3dbd7150506f85292 |
| GPT 5 Mini | 82.6 | Not recorded | 6a1e63b3dbd7150506f85292 |
| SelfCheck | 72.2 | Not recorded | 6a1e63b3dbd7150506f85292 |
| SelfCheck | 67.1 | Not recorded | 6a1e63b3dbd7150506f85292 |
| Llama 4 Scout | 61.3 | Not recorded | 6a1e63b3dbd7150506f85292 |
| Llama 4 Scout | 61.2 | Not recorded | 6a1e63b3dbd7150506f85292 |
| Llama 4 Scout | 60.2 | Not recorded | 6a1e63b3dbd7150506f85292 |
| Random | 58.3 | Not recorded | 6a1e63b3dbd7150506f85292 |

### KG-Guard

F1 Score: 87.4. Result ID: 6a1e8cd7fbc373592c187c20.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

Recorded evaluation configuration: {"Model Backbone":"GraphTransformer"}. This record does not assert independent reproduction.

### KG-Guard

F1 Score: 86.3. Result ID: 6a1e8cd7fbc373592c187c21.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

Recorded evaluation configuration: {"Model Backbone":"GAT"}. This record does not assert independent reproduction.

### GPT 5 Mini

F1 Score: 84.2. Result ID: 6a1e8cd7fbc373592c187c22.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

[Method origin paper](<https://www.sota2.com/research/paper/695f39f053fdd5c361743ecb.md>)

Recorded evaluation configuration: {"Input Context":"G+RS+RT"}. This record does not assert independent reproduction.

### Most Frequent

F1 Score: 83.5. Result ID: 6a1e8cd7fbc373592c187c23.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

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

### GPT 5 Mini

F1 Score: 82.6. Result ID: 6a1e8cd7fbc373592c187c24.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

[Method origin paper](<https://www.sota2.com/research/paper/695f39f053fdd5c361743ecb.md>)

Recorded evaluation configuration: {"Input Context":"G"}. This record does not assert independent reproduction.

### SelfCheck

F1 Score: 72.2. Result ID: 6a1e8cd7fbc373592c187c25.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d943.md>)

Recorded evaluation configuration: {"Sampling size (N)":"10"}. This record does not assert independent reproduction.

### SelfCheck

F1 Score: 67.1. Result ID: 6a1e8cd7fbc373592c187c26.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d943.md>)

Recorded evaluation configuration: {"Sampling size (N)":"5"}. This record does not assert independent reproduction.

### Llama 4 Scout

F1 Score: 61.3. Result ID: 6a1e8cd7fbc373592c187c27.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

Recorded evaluation configuration: {"Input Context":"G+RS"}. This record does not assert independent reproduction.

### Llama 4 Scout

F1 Score: 61.2. Result ID: 6a1e8cd7fbc373592c187c28.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

Recorded evaluation configuration: {"Input Context":"G+RS+RT"}. This record does not assert independent reproduction.

### Llama 4 Scout

F1 Score: 60.2. Result ID: 6a1e8cd7fbc373592c187c29.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

Recorded evaluation configuration: {"Input Context":"G"}. This record does not assert independent reproduction.

### Random

F1 Score: 58.3. Result ID: 6a1e8cd7fbc373592c187c2a.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.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.
