# Hallucination Detection (AUCPR) on SC-GPT4

Source: [Hallucination Detection (AUCPR) on SC-GPT4](<https://www.sota2.com/research/sota/hallucination-detection-aucpr-on-sc-gpt4>)

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

Data updated: 2026-04-08T06:36:52.840Z

Evaluation of model performance on hallucination detection using the SC-GPT4 dataset, specifically measured using AUCPR.

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) | SC-GPT4 |
| Metric | AUCPR |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 18 of 18 |
| 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

- [AUCPR](<https://www.sota2.com/research/sota/hallucination-detection-aucpr-on-sc-gpt4.md?metric=AUCPR>): Higher is better; unit: PERCENT.

## Recorded results

| Method | AUCPR | Result date | Report paper |
| --- | --- | --- | --- |
| SELFCHECKGPT | 95.3 | Not recorded | 68fa7a4a1e209131cea823a4 |
| SELFCHECKGPT | 94.7 | Not recorded | 68fa7a4a1e209131cea823a4 |
| SELFCHECKGPT | 94.4 | Not recorded | 68fa7a4a1e209131cea823a4 |
| SC | 94.2 | Not recorded | 68fa7a4a1e209131cea823a4 |
| SELFCHECKGPT | 93.8 | Not recorded | 68fa7a4a1e209131cea823a4 |
| SELFCHECKGPT | 93.3 | Not recorded | 68fa7a4a1e209131cea823a4 |
| DIRECT | 93 | Not recorded | 68fa7a4a1e209131cea823a4 |
| SELFCHECKGPT | 92.6 | Not recorded | 68fa7a4a1e209131cea823a4 |
| DIRECT | 92.4 | Not recorded | 68fa7a4a1e209131cea823a4 |
| DIRECT | 92.4 | Not recorded | 68fa7a4a1e209131cea823a4 |
| SC | 91.7 | Not recorded | 68fa7a4a1e209131cea823a4 |
| SC | 91.7 | Not recorded | 68fa7a4a1e209131cea823a4 |
| DIRECT | 91.1 | Not recorded | 68fa7a4a1e209131cea823a4 |
| SC | 90.8 | Not recorded | 68fa7a4a1e209131cea823a4 |
| SC | 90.6 | Not recorded | 68fa7a4a1e209131cea823a4 |
| DIRECT | 89.5 | Not recorded | 68fa7a4a1e209131cea823a4 |
| DIRECT | 88.8 | Not recorded | 68fa7a4a1e209131cea823a4 |
| SC | 88.2 | Not recorded | 68fa7a4a1e209131cea823a4 |

### SELFCHECKGPT

AUCPR: 95.3. Result ID: 69d5f6e047f34ce924bdf813.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"LLaMA 3.1","Future context":"true"}. This record does not assert independent reproduction.

### SELFCHECKGPT

AUCPR: 94.7. Result ID: 69d5f6e047f34ce924bdf812.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"LLaMA 3.1","Future context":"false"}. This record does not assert independent reproduction.

### SELFCHECKGPT

AUCPR: 94.4. Result ID: 69d5f6e047f34ce924bdf819.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"Gemma 3","Future context":"true"}. This record does not assert independent reproduction.

### SC

AUCPR: 94.2. Result ID: 69d5f6e047f34ce924bdf815.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"LLaMA 3.1","Future context":"true"}. This record does not assert independent reproduction.

### SELFCHECKGPT

AUCPR: 93.8. Result ID: 69d5f6e047f34ce924bdf818.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"Gemma 3","Future context":"false"}. This record does not assert independent reproduction.

### SELFCHECKGPT

AUCPR: 93.3. Result ID: 69d5f6e047f34ce924bdf81f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"Qwen 2.5","Future context":"true"}. This record does not assert independent reproduction.

### DIRECT

AUCPR: 93. Result ID: 69d5f6e047f34ce924bdf81d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"Qwen 2.5","Future context":"true"}. This record does not assert independent reproduction.

### SELFCHECKGPT

AUCPR: 92.6. Result ID: 69d5f6e047f34ce924bdf81e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"Qwen 2.5","Future context":"false"}. This record does not assert independent reproduction.

### DIRECT

AUCPR: 92.4. Result ID: 69d5f6e047f34ce924bdf811.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"LLaMA 3.1","Future context":"true"}. This record does not assert independent reproduction.

### DIRECT

AUCPR: 92.4. Result ID: 69d5f6e047f34ce924bdf81c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"Qwen 2.5","Future context":"false"}. This record does not assert independent reproduction.

### SC

AUCPR: 91.7. Result ID: 69d5f6e047f34ce924bdf814.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"LLaMA 3.1","Future context":"false"}. This record does not assert independent reproduction.

### SC

AUCPR: 91.7. Result ID: 69d5f6e047f34ce924bdf81b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"Gemma 3","Future context":"true"}. This record does not assert independent reproduction.

### DIRECT

AUCPR: 91.1. Result ID: 69d5f6e047f34ce924bdf810.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"LLaMA 3.1","Future context":"false"}. This record does not assert independent reproduction.

### SC

AUCPR: 90.8. Result ID: 69d5f6e047f34ce924bdf81a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"Gemma 3","Future context":"false"}. This record does not assert independent reproduction.

### SC

AUCPR: 90.6. Result ID: 69d5f6e047f34ce924bdf821.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"Qwen 2.5","Future context":"true"}. This record does not assert independent reproduction.

### DIRECT

AUCPR: 89.5. Result ID: 69d5f6e047f34ce924bdf817.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"Gemma 3","Future context":"true"}. This record does not assert independent reproduction.

### DIRECT

AUCPR: 88.8. Result ID: 69d5f6e047f34ce924bdf816.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"Gemma 3","Future context":"false"}. This record does not assert independent reproduction.

### SC

AUCPR: 88.2. Result ID: 69d5f6e047f34ce924bdf820.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4a1e209131cea823a4.md>)

Recorded evaluation configuration: {"Detector":"Qwen 2.5","Future context":"false"}. This record does not assert independent reproduction.

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