# OOD Detection on Penn Treebank (PTB) Words (test)

Source: [OOD Detection on Penn Treebank (PTB) Words (test)](<https://www.sota2.com/research/sota/ood-detection-on-penn-treebank-ptb-words-test>)

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

Data updated: 2026-02-26T05:44:55.520Z

Evaluation of Out-of-Distribution (OOD) Detection performance on the test split of the Penn Treebank (PTB) Words corpus, typically measured using AUROC, AUPR, and FPR90.

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 | OOD Detection |
| Dataset (including any stated split) | Penn Treebank (PTB) Words (test) |
| Metric | FPR@90 |
| Unit | PERCENT |
| Direction | Lower is better |
| Results loaded | 2 of 2 |
| 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

- [FPR@90](<https://www.sota2.com/research/sota/ood-detection-on-penn-treebank-ptb-words-test.md?metric=FPR%4090>): Lower is better; unit: PERCENT.

- [AUROC](<https://www.sota2.com/research/sota/ood-detection-on-penn-treebank-ptb-words-test.md?metric=AUROC>): Higher is better; unit: PERCENT.

- [AUPR](<https://www.sota2.com/research/sota/ood-detection-on-penn-treebank-ptb-words-test.md?metric=AUPR>): Higher is better; unit: PERCENT.

## Recorded results

| Method | FPR@90 | Result date | Report paper |
| --- | --- | --- | --- |
| Outlier Exposure | 0.98 | Not recorded | 68fa654c1e209131cea096af |
| BPW baseline | 48.5 | Not recorded | 68fa654c1e209131cea096af |

### Outlier Exposure

FPR@90: 0.98. Result ID: 698ca5353ce6cd2163e10f8c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa654c1e209131cea096af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa654c1e209131cea096af.md>)

[Implementation](<https://github.com/hendrycks/outlier-exposure>)

Recorded evaluation configuration: {"scoring\_metric":"Bits Per Word (BPW)","outlier\_exposure":true}. This record does not assert independent reproduction.

### BPW baseline

FPR@90: 48.5. Result ID: 698ca5353ce6cd2163e10f8b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa654c1e209131cea096af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa638d1e209131ce9ff847.md>)

Recorded evaluation configuration: {"scoring\_metric":"Bits Per Word (BPW)","outlier\_exposure":false}. This record does not assert independent reproduction.

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