# Natural Language Understanding on Aggregate NLP Suite 7 Benchmarks

Source: [Natural Language Understanding on Aggregate NLP Suite 7 Benchmarks](<https://www.sota2.com/research/sota/natural-language-understanding-on-aggregate-nlp-suite-7-benchmarks>)

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

Data updated: 2026-03-26T14:17:39.107Z

Evaluation of Natural Language Understanding performance aggregated across the 7 benchmarks contained within the Aggregate NLP Suite.

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) | Aggregate NLP Suite 7 Benchmarks |
| Metric | Average Accuracy |
| Unit | PERCENT |
| 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

- [Average Accuracy](<https://www.sota2.com/research/sota/natural-language-understanding-on-aggregate-nlp-suite-7-benchmarks.md?metric=Average%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Average Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| No pruning | 72.5 | Not recorded | 69c4bdaf0636e5bf4afc76d5 |
| Magnitude-Dim | 65.9 | Not recorded | 69c4bdaf0636e5bf4afc76d5 |
| DIET | 64.6 | Not recorded | 69c4bdaf0636e5bf4afc76d5 |
| DIET | 51.6 | Not recorded | 69c4bdaf0636e5bf4afc76d5 |
| Magnitude-Dim | 51 | Not recorded | 69c4bdaf0636e5bf4afc76d5 |
| PuDDing | 49.7 | Not recorded | 69c4bdaf0636e5bf4afc76d5 |
| PuDDing | 48.9 | Not recorded | 69c4bdaf0636e5bf4afc76d5 |
| SliceGPT | 35.3 | Not recorded | 69c4bdaf0636e5bf4afc76d5 |
| SliceGPT | 35.2 | Not recorded | 69c4bdaf0636e5bf4afc76d5 |

### No pruning

Average Accuracy: 72.5. Result ID: 69c53f7e47f34ce924af122f.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdaf0636e5bf4afc76d5.md>)

Recorded evaluation configuration: {"Sparsity":"0","Backbone":"Gemma-2 9B","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

### Magnitude-Dim

Average Accuracy: 65.9. Result ID: 69c53f7e47f34ce924af1234.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdaf0636e5bf4afc76d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa622d1e209131ce9f7f15.md>)

Recorded evaluation configuration: {"Sparsity":"10%","Backbone":"Gemma-2 9B","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

### DIET

Average Accuracy: 64.6. Result ID: 69c53f7e47f34ce924af1237.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdaf0636e5bf4afc76d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c4bdaf0636e5bf4afc76d5.md>)

Recorded evaluation configuration: {"Sparsity":"10%","Backbone":"Gemma-2 9B","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

### DIET

Average Accuracy: 51.6. Result ID: 69c53f7e47f34ce924af1233.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdaf0636e5bf4afc76d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c4bdaf0636e5bf4afc76d5.md>)

Recorded evaluation configuration: {"Sparsity":"20%","Backbone":"Gemma-2 9B","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

### Magnitude-Dim

Average Accuracy: 51. Result ID: 69c53f7e47f34ce924af1230.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdaf0636e5bf4afc76d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa622d1e209131ce9f7f15.md>)

Recorded evaluation configuration: {"Sparsity":"20%","Backbone":"Gemma-2 9B","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

### PuDDing

Average Accuracy: 49.7. Result ID: 69c53f7e47f34ce924af1236.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdaf0636e5bf4afc76d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa794f1e209131cea7d0aa.md>)

Recorded evaluation configuration: {"Sparsity":"10%","Backbone":"Gemma-2 9B","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

### PuDDing

Average Accuracy: 48.9. Result ID: 69c53f7e47f34ce924af1232.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdaf0636e5bf4afc76d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa794f1e209131cea7d0aa.md>)

Recorded evaluation configuration: {"Sparsity":"20%","Backbone":"Gemma-2 9B","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

### SliceGPT

Average Accuracy: 35.3. Result ID: 69c53f7e47f34ce924af1231.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdaf0636e5bf4afc76d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72271e209131cea54d13.md>)

Recorded evaluation configuration: {"Sparsity":"20%","Backbone":"Gemma-2 9B","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

### SliceGPT

Average Accuracy: 35.2. Result ID: 69c53f7e47f34ce924af1235.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdaf0636e5bf4afc76d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72271e209131cea54d13.md>)

Recorded evaluation configuration: {"Sparsity":"10%","Backbone":"Gemma-2 9B","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

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