# Language Modeling on WikiText-103 5.0% corruption (test)

Source: [Language Modeling on WikiText-103 5.0% corruption (test)](<https://www.sota2.com/research/sota/language-modeling-on-wikitext-103-5-0-percent-corruption-test>)

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

Data updated: 2026-04-02T08:43:04.554Z

Evaluation of language modeling performance using the 5.0% corrupted test split of the WikiText-103 dataset, typically measured by Perplexity (PPL).

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 | Language Modeling |
| Dataset (including any stated split) | WikiText-103 5.0% corruption (test) |
| Metric | Perplexity (PPL) |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 8 of 8 |
| 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

- [Perplexity (PPL)](<https://www.sota2.com/research/sota/language-modeling-on-wikitext-103-5-0-percent-corruption-test.md?metric=Perplexity%20(PPL)>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Perplexity (PPL) | Result date | Report paper |
| --- | --- | --- | --- |
| Transformer-XL | 43.511 | Not recorded | 69cdf7f20636e5bf4afcc99a |
| Transformer-XL | 44.168 | Not recorded | 69cdf7f20636e5bf4afcc99a |
| Transformer | 54.531 | Not recorded | 69cdf7f20636e5bf4afcc99a |
| Transformer | 54.899 | Not recorded | 69cdf7f20636e5bf4afcc99a |
| Transformer | 55.012 | Not recorded | 69cdf7f20636e5bf4afcc99a |
| Transformer | 55.451 | Not recorded | 69cdf7f20636e5bf4afcc99a |
| Transformer | 60.944 | Not recorded | 69cdf7f20636e5bf4afcc99a |
| Transformer | 61.062 | Not recorded | 69cdf7f20636e5bf4afcc99a |

### Transformer-XL

Perplexity (PPL): 43.511. Result ID: 69ce2bef47f34ce924b75d63.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f20636e5bf4afcc99a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf7f20636e5bf4afcc99a.md>)

Recorded evaluation configuration: {"Model size (Parameters)":"Base (151M)","Attention":"QUEST"}. This record does not assert independent reproduction.

### Transformer-XL

Perplexity (PPL): 44.168. Result ID: 69ce2bef47f34ce924b75d62.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f20636e5bf4afcc99a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf7f20636e5bf4afcc99a.md>)

Recorded evaluation configuration: {"Model size (Parameters)":"Base (151M)","Attention":"Standard"}. This record does not assert independent reproduction.

### Transformer

Perplexity (PPL): 54.531. Result ID: 69ce2bef47f34ce924b75d61.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f20636e5bf4afcc99a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf7f20636e5bf4afcc99a.md>)

Recorded evaluation configuration: {"Model size (Parameters)":"Medium (90M)","Attention":"QUEST"}. This record does not assert independent reproduction.

### Transformer

Perplexity (PPL): 54.899. Result ID: 69ce2bef47f34ce924b75d60.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f20636e5bf4afcc99a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf7f20636e5bf4afcc99a.md>)

Recorded evaluation configuration: {"Model size (Parameters)":"Medium (90M)","Attention":"QKNorm-DS"}. This record does not assert independent reproduction.

### Transformer

Perplexity (PPL): 55.012. Result ID: 69ce2bef47f34ce924b75d5e.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f20636e5bf4afcc99a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf7f20636e5bf4afcc99a.md>)

Recorded evaluation configuration: {"Model size (Parameters)":"Medium (90M)","Attention":"Standard"}. This record does not assert independent reproduction.

### Transformer

Perplexity (PPL): 55.451. Result ID: 69ce2bef47f34ce924b75d5f.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f20636e5bf4afcc99a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf7f20636e5bf4afcc99a.md>)

Recorded evaluation configuration: {"Model size (Parameters)":"Medium (90M)","Attention":"QNorm"}. This record does not assert independent reproduction.

### Transformer

Perplexity (PPL): 60.944. Result ID: 69ce2bef47f34ce924b75d5c.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f20636e5bf4afcc99a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf7f20636e5bf4afcc99a.md>)

Recorded evaluation configuration: {"Model size (Parameters)":"Small (44M)","Attention":"Standard"}. This record does not assert independent reproduction.

### Transformer

Perplexity (PPL): 61.062. Result ID: 69ce2bef47f34ce924b75d5d.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf7f20636e5bf4afcc99a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf7f20636e5bf4afcc99a.md>)

Recorded evaluation configuration: {"Model size (Parameters)":"Small (44M)","Attention":"QUEST"}. This record does not assert independent reproduction.

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