# Uncertainty Estimation on Natural Questions (NQ)

Source: [Uncertainty Estimation on Natural Questions (NQ)](<https://www.sota2.com/research/sota/uncertainty-estimation-on-natural-questions-nq>)

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

Data updated: 2026-05-01T08:43:34.562Z

Evaluation of model uncertainty estimation performance on the Natural Questions (NQ) dataset, reporting metrics such as AUROC and AUPRC.

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 | Uncertainty Estimation |
| Dataset (including any stated split) | Natural Questions (NQ) |
| Metric | AUROC |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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/uncertainty-estimation-on-natural-questions-nq.md?metric=AUROC>): Higher is better; unit: SCALAR.

- [AUPRC](<https://www.sota2.com/research/sota/uncertainty-estimation-on-natural-questions-nq.md?metric=AUPRC>): Higher is better; unit: SCALAR.

## Recorded results

| Method | AUROC | Result date | Report paper |
| --- | --- | --- | --- |
| Proposed | 0.77 | Not recorded | 69f4339755a3c32af89fd154 |
| SAR | 0.75 | Not recorded | 69f4339755a3c32af89fd154 |
| Perplexity | 0.67 | Not recorded | 69f4339755a3c32af89fd154 |
| Focus | 0.64 | Not recorded | 69f4339755a3c32af89fd154 |
| Semantic | 0.63 | Not recorded | 69f4339755a3c32af89fd154 |
| P(True) | 0.61 | Not recorded | 69f4339755a3c32af89fd154 |
| Eigen | 0.57 | Not recorded | 69f4339755a3c32af89fd154 |
| ATRMD | 0.54 | Not recorded | 69f4339755a3c32af89fd154 |
| Attention | 0.49 | Not recorded | 69f4339755a3c32af89fd154 |

### Proposed

AUROC: 0.77. Result ID: 69f466de47f34ce924da6b21.

[Reporting paper](<https://www.sota2.com/research/paper/69f4339755a3c32af89fd154.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f4339755a3c32af89fd154.md>)

Recorded evaluation configuration: {"Backbone Model":"LLaMA-3-8B-Instruct"}. This record does not assert independent reproduction.

### SAR

AUROC: 0.75. Result ID: 69f466de47f34ce924da6b1a.

[Reporting paper](<https://www.sota2.com/research/paper/69f4339755a3c32af89fd154.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73681e209131cea5c472.md>)

Recorded evaluation configuration: {"Backbone Model":"LLaMA-3-8B-Instruct"}. This record does not assert independent reproduction.

### Perplexity

AUROC: 0.67. Result ID: 69f466de47f34ce924da6b19.

[Reporting paper](<https://www.sota2.com/research/paper/69f4339755a3c32af89fd154.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67f11e209131cea19446.md>)

Recorded evaluation configuration: {"Backbone Model":"LLaMA-3-8B-Instruct"}. This record does not assert independent reproduction.

### Focus

AUROC: 0.64. Result ID: 69f466de47f34ce924da6b1c.

[Reporting paper](<https://www.sota2.com/research/paper/69f4339755a3c32af89fd154.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa713d1e209131cea50031.md>)

Recorded evaluation configuration: {"Backbone Model":"LLaMA-3-8B-Instruct"}. This record does not assert independent reproduction.

### Semantic

AUROC: 0.63. Result ID: 69f466de47f34ce924da6b1b.

[Reporting paper](<https://www.sota2.com/research/paper/69f4339755a3c32af89fd154.md>)

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

Recorded evaluation configuration: {"Backbone Model":"LLaMA-3-8B-Instruct"}. This record does not assert independent reproduction.

### P(True)

AUROC: 0.61. Result ID: 69f466de47f34ce924da6b20.

[Reporting paper](<https://www.sota2.com/research/paper/69f4339755a3c32af89fd154.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6df71e209131cea3d034.md>)

Recorded evaluation configuration: {"Backbone Model":"LLaMA-3-8B-Instruct"}. This record does not assert independent reproduction.

### Eigen

AUROC: 0.57. Result ID: 69f466de47f34ce924da6b1e.

[Reporting paper](<https://www.sota2.com/research/paper/69f4339755a3c32af89fd154.md>)

Recorded evaluation configuration: {"Backbone Model":"LLaMA-3-8B-Instruct"}. This record does not assert independent reproduction.

### ATRMD

AUROC: 0.54. Result ID: 69f466de47f34ce924da6b1f.

[Reporting paper](<https://www.sota2.com/research/paper/69f4339755a3c32af89fd154.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa781c1e209131cea76d53.md>)

Recorded evaluation configuration: {"Backbone Model":"LLaMA-3-8B-Instruct"}. This record does not assert independent reproduction.

### Attention

AUROC: 0.49. Result ID: 69f466de47f34ce924da6b1d.

[Reporting paper](<https://www.sota2.com/research/paper/69f4339755a3c32af89fd154.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67225.md>)

Recorded evaluation configuration: {"Backbone Model":"LLaMA-3-8B-Instruct"}. This record does not assert independent reproduction.

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