# Natural Language Understanding on WG, Yelp, SciTail, and PAWS Aggregate

Source: [Natural Language Understanding on WG, Yelp, SciTail, and PAWS Aggregate](<https://www.sota2.com/research/sota/natural-language-understanding-on-wg-yelp-scitail-and-paws-aggregate>)

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

Data updated: 2026-02-26T06:08:39.176Z

Aggregate evaluation of Natural Language Understanding performance across the WG, Yelp, SciTail, and PAWS datasets/tasks.

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 | Natural Language Understanding |
| Dataset (including any stated split) | Others (WG, Yelp, SciTail, PAWS) |
| Metric | WG Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 11 of 11 |
| 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

- [WG Accuracy](<https://www.sota2.com/research/sota/natural-language-understanding-on-wg-yelp-scitail-and-paws-aggregate.md?metric=WG%20Accuracy>): Higher is better; unit: PERCENT.

- [Yelp Accuracy](<https://www.sota2.com/research/sota/natural-language-understanding-on-wg-yelp-scitail-and-paws-aggregate.md?metric=Yelp%20Accuracy>): Higher is better; unit: PERCENT.

- [SciTail Accuracy](<https://www.sota2.com/research/sota/natural-language-understanding-on-wg-yelp-scitail-and-paws-aggregate.md?metric=SciTail%20Accuracy>): Higher is better; unit: PERCENT.

- [PAWS Accuracy](<https://www.sota2.com/research/sota/natural-language-understanding-on-wg-yelp-scitail-and-paws-aggregate.md?metric=PAWS%20Accuracy>): Higher is better; unit: PERCENT.

- [Aggregate Accuracy](<https://www.sota2.com/research/sota/natural-language-understanding-on-wg-yelp-scitail-and-paws-aggregate.md?metric=Aggregate%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | WG Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| PEMT | 62.3 | Not recorded | 68fa738b1e209131cea5d1bc |
| FT | 61.9 | Not recorded | 68fa738b1e209131cea5d1bc |
| LoRA | 60.2 | Not recorded | 68fa738b1e209131cea5d1bc |
| Adamix | 59.8 | Not recorded | 68fa738b1e209131cea5d1bc |
| Adapter | 59.2 | Not recorded | 68fa738b1e209131cea5d1bc |
| ATTEMPT | 57.6 | Not recorded | 68fa738b1e209131cea5d1bc |
| BitFit | 57.2 | Not recorded | 68fa738b1e209131cea5d1bc |
| MPT | 56.5 | Not recorded | 68fa738b1e209131cea5d1bc |
| MixDA | 55.2 | Not recorded | 68fa738b1e209131cea5d1bc |
| SPoT | 50.4 | Not recorded | 68fa738b1e209131cea5d1bc |
| PT | 49.6 | Not recorded | 68fa738b1e209131cea5d1bc |

### PEMT

WG Accuracy: 62.3. Result ID: 6992ef273a6c3148491276a4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738b1e209131cea5d1bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738b1e209131cea5d1bc.md>)

[Implementation](<https://github.com/JachinLin2022/PEMT>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### FT

WG Accuracy: 61.9. Result ID: 6992ef273a6c31484912769a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738b1e209131cea5d1bc.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### LoRA

WG Accuracy: 60.2. Result ID: 6992ef273a6c31484912769e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738b1e209131cea5d1bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b121e209131cea2bf59.md>)

[Implementation](<https://github.com/microsoft/LoRA>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Adamix

WG Accuracy: 59.8. Result ID: 6992ef273a6c3148491276a3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738b1e209131cea5d1bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c717.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Adapter

WG Accuracy: 59.2. Result ID: 6992ef273a6c31484912769d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738b1e209131cea5d1bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65e31e209131cea0d324.md>)

[Implementation](<https://github.com/google-research/adapter-bert>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### ATTEMPT

WG Accuracy: 57.6. Result ID: 6992ef273a6c3148491276a0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738b1e209131cea5d1bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e221e209131cea3e0f3.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### BitFit

WG Accuracy: 57.2. Result ID: 6992ef273a6c31484912769c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738b1e209131cea5d1bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d5a1e209131cea39632.md>)

[Implementation](<https://www.github.com/benzakenelad/BitFit>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### MPT

WG Accuracy: 56.5. Result ID: 6992ef273a6c3148491276a1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738b1e209131cea5d1bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb51e209131cea41a8d.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### MixDA

WG Accuracy: 55.2. Result ID: 6992ef273a6c3148491276a2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738b1e209131cea5d1bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fb11e209131cea46eba.md>)

[Implementation](<https://github.com/Amano-Aki/Mixture-of-Domain-Adapters>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### SPoT

WG Accuracy: 50.4. Result ID: 6992ef273a6c31484912769f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738b1e209131cea5d1bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bfd1e209131cea31722.md>)

[Implementation](<https://github.com/google-research/prompt-tuning/tree/main/prompt_tuning/spot>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### PT

WG Accuracy: 49.6. Result ID: 6992ef273a6c31484912769b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738b1e209131cea5d1bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6aa51e209131cea296c9.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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