# Hallucination Detection on PopQA n=1000 (test)

Source: [Hallucination Detection on PopQA n=1000 (test)](<https://www.sota2.com/research/sota/hallucination-detection-on-popqa-n-1000-test>)

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

Data updated: 2026-05-07T10:03:59.886Z

Evaluation of model performance on Hallucination Detection using the specific n=1000 test subset of the PopQA dataset.

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) | PopQA n=1000 (test) |
| Metric | AUROC |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 24 of 24 |
| 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

- [AUROC](<https://www.sota2.com/research/sota/hallucination-detection-on-popqa-n-1000-test.md?metric=AUROC>): Higher is better; unit: SCALAR.

- [Delta Score](<https://www.sota2.com/research/sota/hallucination-detection-on-popqa-n-1000-test.md?metric=Delta%20Score>): Higher is better; unit: SCALAR.

- [Accuracy](<https://www.sota2.com/research/sota/hallucination-detection-on-popqa-n-1000-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | AUROC | Result date | Report paper |
| --- | --- | --- | --- |
| phi\_first | 0.895 | Not recorded | 69fc1cc45497b3b670c88788 |
| phi\_first | 0.887 | Not recorded | 69fc1cc45497b3b670c88788 |
| phi\_first | 0.875 | Not recorded | 69fc1cc45497b3b670c88788 |
| Sem. AU | 0.874 | Not recorded | 69fc1cc45497b3b670c88788 |
| Sem. AU | 0.867 | Not recorded | 69fc1cc45497b3b670c88788 |
| AU-full | 0.866 | Not recorded | 69fc1cc45497b3b670c88788 |
| AU-full | 0.857 | Not recorded | 69fc1cc45497b3b670c88788 |
| AU-3w | 0.856 | Not recorded | 69fc1cc45497b3b670c88788 |
| AU-3w | 0.854 | Not recorded | 69fc1cc45497b3b670c88788 |
| phi\_first | 0.842 | Not recorded | 69fc1cc45497b3b670c88788 |
| AU-1w | 0.84 | Not recorded | 69fc1cc45497b3b670c88788 |
| Sem. AU | 0.839 | Not recorded | 69fc1cc45497b3b670c88788 |
| AU-full | 0.834 | Not recorded | 69fc1cc45497b3b670c88788 |
| AU-1w | 0.831 | Not recorded | 69fc1cc45497b3b670c88788 |
| AU-3w | 0.824 | Not recorded | 69fc1cc45497b3b670c88788 |
| AU-1w | 0.802 | Not recorded | 69fc1cc45497b3b670c88788 |
| Verbalized | 0.782 | Not recorded | 69fc1cc45497b3b670c88788 |
| AU-full | 0.778 | Not recorded | 69fc1cc45497b3b670c88788 |
| Sem. AU | 0.775 | Not recorded | 69fc1cc45497b3b670c88788 |
| AU-3w | 0.762 | Not recorded | 69fc1cc45497b3b670c88788 |
| AU-1w | 0.735 | Not recorded | 69fc1cc45497b3b670c88788 |
| Verbalized | 0.705 | Not recorded | 69fc1cc45497b3b670c88788 |
| Verbalized | 0.701 | Not recorded | 69fc1cc45497b3b670c88788 |
| Verbalized | 0.632 | Not recorded | 69fc1cc45497b3b670c88788 |

### phi\_first

AUROC: 0.895. Result ID: 69fc637347f34ce924dfa130.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-7B","Inference Cost":"1 decode"}. This record does not assert independent reproduction.

### phi\_first

AUROC: 0.887. Result ID: 69fc637347f34ce924dfa124.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

Recorded evaluation configuration: {"Model":"Llama-3.1-8B","Inference Cost":"1 decode"}. This record does not assert independent reproduction.

### phi\_first

AUROC: 0.875. Result ID: 69fc637347f34ce924dfa136.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

Recorded evaluation configuration: {"Model":"PopQA mean","Inference Cost":"1 decode"}. This record does not assert independent reproduction.

### Sem. AU

AUROC: 0.874. Result ID: 69fc637347f34ce924dfa123.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f141e209131cea43bf3.md>)

Recorded evaluation configuration: {"Model":"Llama-3.1-8B","Inference Cost":"1+10 decodes (sampling)"}. This record does not assert independent reproduction.

### Sem. AU

AUROC: 0.867. Result ID: 69fc637347f34ce924dfa12f.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f141e209131cea43bf3.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-7B","Inference Cost":"1+10 decodes (sampling)"}. This record does not assert independent reproduction.

### AU-full

AUROC: 0.866. Result ID: 69fc637347f34ce924dfa122.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

Recorded evaluation configuration: {"Model":"Llama-3.1-8B","Inference Cost":"1+10 decodes (sampling)"}. This record does not assert independent reproduction.

### AU-full

AUROC: 0.857. Result ID: 69fc637347f34ce924dfa12e.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-7B","Inference Cost":"1+10 decodes (sampling)"}. This record does not assert independent reproduction.

### AU-3w

AUROC: 0.856. Result ID: 69fc637347f34ce924dfa12d.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-7B","Inference Cost":"1+10 decodes (sampling)"}. This record does not assert independent reproduction.

### AU-3w

AUROC: 0.854. Result ID: 69fc637347f34ce924dfa121.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

Recorded evaluation configuration: {"Model":"Llama-3.1-8B","Inference Cost":"1+10 decodes (sampling)"}. This record does not assert independent reproduction.

### phi\_first

AUROC: 0.842. Result ID: 69fc637347f34ce924dfa12a.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

Recorded evaluation configuration: {"Model":"Mistral-7B-v0.3","Inference Cost":"1 decode"}. This record does not assert independent reproduction.

### AU-1w

AUROC: 0.84. Result ID: 69fc637347f34ce924dfa120.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

Recorded evaluation configuration: {"Model":"Llama-3.1-8B","Inference Cost":"1+10 decodes (sampling)"}. This record does not assert independent reproduction.

### Sem. AU

AUROC: 0.839. Result ID: 69fc637347f34ce924dfa135.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f141e209131cea43bf3.md>)

Recorded evaluation configuration: {"Model":"PopQA mean","Inference Cost":"1+10 decodes (sampling)"}. This record does not assert independent reproduction.

### AU-full

AUROC: 0.834. Result ID: 69fc637347f34ce924dfa134.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

Recorded evaluation configuration: {"Model":"PopQA mean","Inference Cost":"1+10 decodes (sampling)"}. This record does not assert independent reproduction.

### AU-1w

AUROC: 0.831. Result ID: 69fc637347f34ce924dfa12c.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-7B","Inference Cost":"1+10 decodes (sampling)"}. This record does not assert independent reproduction.

### AU-3w

AUROC: 0.824. Result ID: 69fc637347f34ce924dfa133.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

Recorded evaluation configuration: {"Model":"PopQA mean","Inference Cost":"1+10 decodes (sampling)"}. This record does not assert independent reproduction.

### AU-1w

AUROC: 0.802. Result ID: 69fc637347f34ce924dfa132.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

Recorded evaluation configuration: {"Model":"PopQA mean","Inference Cost":"1+10 decodes (sampling)"}. This record does not assert independent reproduction.

### Verbalized

AUROC: 0.782. Result ID: 69fc637347f34ce924dfa12b.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70f81e209131cea4e72e.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-7B","Inference Cost":"1 decode"}. This record does not assert independent reproduction.

### AU-full

AUROC: 0.778. Result ID: 69fc637347f34ce924dfa128.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

Recorded evaluation configuration: {"Model":"Mistral-7B-v0.3","Inference Cost":"1+10 decodes (sampling)"}. This record does not assert independent reproduction.

### Sem. AU

AUROC: 0.775. Result ID: 69fc637347f34ce924dfa129.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f141e209131cea43bf3.md>)

Recorded evaluation configuration: {"Model":"Mistral-7B-v0.3","Inference Cost":"1+10 decodes (sampling)"}. This record does not assert independent reproduction.

### AU-3w

AUROC: 0.762. Result ID: 69fc637347f34ce924dfa127.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

Recorded evaluation configuration: {"Model":"Mistral-7B-v0.3","Inference Cost":"1+10 decodes (sampling)"}. This record does not assert independent reproduction.

### AU-1w

AUROC: 0.735. Result ID: 69fc637347f34ce924dfa126.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

Recorded evaluation configuration: {"Model":"Mistral-7B-v0.3","Inference Cost":"1+10 decodes (sampling)"}. This record does not assert independent reproduction.

### Verbalized

AUROC: 0.705. Result ID: 69fc637347f34ce924dfa131.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70f81e209131cea4e72e.md>)

Recorded evaluation configuration: {"Model":"PopQA mean","Inference Cost":"1 decode"}. This record does not assert independent reproduction.

### Verbalized

AUROC: 0.701. Result ID: 69fc637347f34ce924dfa125.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70f81e209131cea4e72e.md>)

Recorded evaluation configuration: {"Model":"Mistral-7B-v0.3","Inference Cost":"1 decode"}. This record does not assert independent reproduction.

### Verbalized

AUROC: 0.632. Result ID: 69fc637347f34ce924dfa11f.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1cc45497b3b670c88788.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70f81e209131cea4e72e.md>)

Recorded evaluation configuration: {"Model":"Llama-3.1-8B","Inference Cost":"1 decode"}. This record does not assert independent reproduction.

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