# Pronoun Disambiguation on WSC (test)

Source: [Pronoun Disambiguation on WSC (test)](<https://www.sota2.com/research/sota/pronoun-disambiguation-on-wsc-test>)

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

Data updated: 2026-02-26T05:34:26.961Z

Evaluation of model performance on resolving pronoun ambiguities using the test split of the Winograd Schema Challenge (WSC) dataset.

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 | Pronoun Disambiguation |
| Dataset (including any stated split) | WSC (test) |
| Metric | Accuracy (Single) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 24 of 24 |
| 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

- [Accuracy (Single)](<https://www.sota2.com/research/sota/pronoun-disambiguation-on-wsc-test.md?metric=Accuracy%20(Single)>): Higher is better; unit: PERCENT.

- [Overall Accuracy](<https://www.sota2.com/research/sota/pronoun-disambiguation-on-wsc-test.md?metric=Overall%20Accuracy>): Higher is better; unit: PERCENT.

- [Accuracy (Non-Associated)](<https://www.sota2.com/research/sota/pronoun-disambiguation-on-wsc-test.md?metric=Accuracy%20(Non-Associated)>): Higher is better; unit: PERCENT.

- [Accuracy (Associated)](<https://www.sota2.com/research/sota/pronoun-disambiguation-on-wsc-test.md?metric=Accuracy%20(Associated)>): Higher is better; unit: PERCENT.

- [Accuracy (Unswitched)](<https://www.sota2.com/research/sota/pronoun-disambiguation-on-wsc-test.md?metric=Accuracy%20(Unswitched)>): Higher is better; unit: PERCENT.

- [Accuracy (Switched)](<https://www.sota2.com/research/sota/pronoun-disambiguation-on-wsc-test.md?metric=Accuracy%20(Switched)>): Higher is better; unit: PERCENT.

- [Consistency](<https://www.sota2.com/research/sota/pronoun-disambiguation-on-wsc-test.md?metric=Consistency>): Higher is better; unit: PERCENT.

- [Accuracy (Group)](<https://www.sota2.com/research/sota/pronoun-disambiguation-on-wsc-test.md?metric=Accuracy%20(Group)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (Single) | Result date | Report paper |
| --- | --- | --- | --- |
| ALBERT-xxlarge | 78.8 | Not recorded | 68fa6b131e209131cea2bf75 |
| ALBERT-xxlarge | 77.4 | Not recorded | 68fa6b131e209131cea2bf75 |
| RoBERTa-large | 73.91 | Not recorded | 68fa6b131e209131cea2bf75 |
| RoBERTa-large | 71.23 | Not recorded | 68fa6b131e209131cea2bf75 |
| RoBERTa-base | 63.04 | Not recorded | 68fa6b131e209131cea2bf75 |
| BERT-large | 61.41 | Not recorded | 68fa6b131e209131cea2bf75 |
| BERT-large | 60.27 | Not recorded | 68fa6b131e209131cea2bf75 |
| RoBERTa-base | 60.27 | Not recorded | 68fa6b131e209131cea2bf75 |
| BERT-base | 56.52 | Not recorded | 68fa6b131e209131cea2bf75 |
| ALBERT-base | 55.48 | Not recorded | 68fa6b131e209131cea2bf75 |
| ALBERT-base | 55.43 | Not recorded | 68fa6b131e209131cea2bf75 |
| BERT-base | 54.79 | Not recorded | 68fa6b131e209131cea2bf75 |
| random | 50 | Not recorded | 68fa6b131e209131cea2bf75 |
| random | 50 | Not recorded | 68fa6b131e209131cea2bf75 |

### ALBERT-xxlarge

Accuracy (Single): 78.8. Result ID: 698e5296ffc3901de09c9da7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b131e209131cea2bf75.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67061e209131cea13e88.md>)

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

Recorded evaluation configuration: {"Setup":"Zero-shot"}. This record does not assert independent reproduction.

### ALBERT-xxlarge

Accuracy (Single): 77.4. Result ID: 698e5296ffc3901de09c9dae.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b131e209131cea2bf75.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67061e209131cea13e88.md>)

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

Recorded evaluation configuration: {"Setup":"Zero-shot"}. This record does not assert independent reproduction.

### RoBERTa-large

Accuracy (Single): 73.91. Result ID: 698e5296ffc3901de09c9da5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b131e209131cea2bf75.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66131e209131cea0e18b.md>)

[Implementation](<https://github.com/pytorch/fairseq>)

Recorded evaluation configuration: {"Setup":"Zero-shot"}. This record does not assert independent reproduction.

### RoBERTa-large

Accuracy (Single): 71.23. Result ID: 698e5296ffc3901de09c9dac.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b131e209131cea2bf75.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66131e209131cea0e18b.md>)

[Implementation](<https://github.com/pytorch/fairseq>)

Recorded evaluation configuration: {"Setup":"Zero-shot"}. This record does not assert independent reproduction.

### RoBERTa-base

Accuracy (Single): 63.04. Result ID: 698e5296ffc3901de09c9da4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b131e209131cea2bf75.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66131e209131cea0e18b.md>)

[Implementation](<https://github.com/pytorch/fairseq>)

Recorded evaluation configuration: {"Setup":"Zero-shot"}. This record does not assert independent reproduction.

### BERT-large

Accuracy (Single): 61.41. Result ID: 698e5296ffc3901de09c9da3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b131e209131cea2bf75.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c31e209131cea0c4ed.md>)

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

Recorded evaluation configuration: {"Setup":"Zero-shot"}. This record does not assert independent reproduction.

### BERT-large

Accuracy (Single): 60.27. Result ID: 698e5296ffc3901de09c9daa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b131e209131cea2bf75.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c31e209131cea0c4ed.md>)

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

Recorded evaluation configuration: {"Setup":"Zero-shot"}. This record does not assert independent reproduction.

### RoBERTa-base

Accuracy (Single): 60.27. Result ID: 698e5296ffc3901de09c9dab.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b131e209131cea2bf75.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66131e209131cea0e18b.md>)

[Implementation](<https://github.com/pytorch/fairseq>)

Recorded evaluation configuration: {"Setup":"Zero-shot"}. This record does not assert independent reproduction.

### BERT-base

Accuracy (Single): 56.52. Result ID: 698e5296ffc3901de09c9da2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b131e209131cea2bf75.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c31e209131cea0c4ed.md>)

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

Recorded evaluation configuration: {"Setup":"Zero-shot"}. This record does not assert independent reproduction.

### ALBERT-base

Accuracy (Single): 55.48. Result ID: 698e5296ffc3901de09c9dad.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b131e209131cea2bf75.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67061e209131cea13e88.md>)

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

Recorded evaluation configuration: {"Setup":"Zero-shot"}. This record does not assert independent reproduction.

### ALBERT-base

Accuracy (Single): 55.43. Result ID: 698e5296ffc3901de09c9da6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b131e209131cea2bf75.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67061e209131cea13e88.md>)

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

Recorded evaluation configuration: {"Setup":"Zero-shot"}. This record does not assert independent reproduction.

### BERT-base

Accuracy (Single): 54.79. Result ID: 698e5296ffc3901de09c9da9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b131e209131cea2bf75.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c31e209131cea0c4ed.md>)

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

Recorded evaluation configuration: {"Setup":"Zero-shot"}. This record does not assert independent reproduction.

### random

Accuracy (Single): 50. Result ID: 698e5296ffc3901de09c9da1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b131e209131cea2bf75.md>)

Recorded evaluation configuration: {"Setup":"Zero-shot"}. This record does not assert independent reproduction.

### random

Accuracy (Single): 50. Result ID: 698e5296ffc3901de09c9da8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b131e209131cea2bf75.md>)

Recorded evaluation configuration: {"Setup":"Zero-shot"}. This record does not assert independent reproduction.

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