# Reading Comprehension on Who Did What (WDW) (test)

Source: [Reading Comprehension on Who Did What (WDW) (test)](<https://www.sota2.com/research/sota/reading-comprehension-on-who-did-what-wdw-test>)

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

Data updated: 2026-02-26T05:27:16.087Z

Evaluation of model performance on the test split of the Who Did What (WDW) dataset for Reading Comprehension tasks, typically measured using Accuracy.

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 | Reading Comprehension |
| Dataset (including any stated split) | Who Did What (WDW) (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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](<https://www.sota2.com/research/sota/reading-comprehension-on-who-did-what-wdw-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Fine-Grained Gating | 71.7 | Not recorded | 68fa63a41e209131ce9fff99 |
| Gated-Attention (GA) Reader | 71.2 | Not recorded | 68fa63a41e209131ce9fff99 |
| Munkhdalai & Yu (2016) | 66.2 | Not recorded | 68fa63a41e209131ce9fff99 |
| Chen et al. (2016) | 64 | Not recorded | 68fa63a41e209131ce9fff99 |
| Kadlec et al. (2016) | 57 | Not recorded | 68fa63a41e209131ce9fff99 |

### Fine-Grained Gating

Accuracy: 71.7. Result ID: 698bfe57f729bc7f68e7651e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa63a41e209131ce9fff99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63a41e209131ce9fff99.md>)

[Implementation](<https://github.com/kimiyoung/fg-gating>)

Recorded evaluation configuration: {"word embeddings":"fixed"}. This record does not assert independent reproduction.

### Gated-Attention (GA) Reader

Accuracy: 71.2. Result ID: 698bfe57f729bc7f68e7651d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa63a41e209131ce9fff99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63491e209131ce9fdb44.md>)

[Implementation](<https://github.com/bdhingra/ga-reader>)

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

### Munkhdalai & Yu (2016)

Accuracy: 66.2. Result ID: 698bfe57f729bc7f68e7651c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa63a41e209131ce9fff99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63051e209131ce9fc644.md>)

[Implementation](<https://bitbucket.org/tsendeemts/nse>)

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

### Chen et al. (2016)

Accuracy: 64. Result ID: 698bfe57f729bc7f68e7651b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa63a41e209131ce9fff99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62a21e209131ce9fa624.md>)

[Implementation](<https://github.com/danqi/rc-cnn-dailymail>)

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

### Kadlec et al. (2016)

Accuracy: 57. Result ID: 698bfe57f729bc7f68e7651a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa63a41e209131ce9fff99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa629e1e209131ce9fa3d7.md>)

[Implementation](<https://github.com/rkadlec/asreader>)

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.