# Supertagging on French TLGbank (test)

Source: [Supertagging on French TLGbank (test)](<https://www.sota2.com/research/sota/supertagging-on-french-tlgbank-test>)

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

Data updated: 2026-02-26T05:54:24.103Z

Evaluation of Supertagging performance on the test split of the French TLGbank dataset, using frequency-based accuracy metrics.

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 | Supertagging |
| Dataset (including any stated split) | French TLGbank (test) |
| Metric | Overall Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

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

- [Frequent Accuracy](<https://www.sota2.com/research/sota/supertagging-on-french-tlgbank-test.md?metric=Frequent%20Accuracy>): Higher is better; unit: PERCENT.

- [Uncommon Accuracy](<https://www.sota2.com/research/sota/supertagging-on-french-tlgbank-test.md?metric=Uncommon%20Accuracy>): Higher is better; unit: PERCENT.

- [Rare Accuracy](<https://www.sota2.com/research/sota/supertagging-on-french-tlgbank-test.md?metric=Rare%20Accuracy>): Higher is better; unit: PERCENT.

- [Unseen Accuracy](<https://www.sota2.com/research/sota/supertagging-on-french-tlgbank-test.md?metric=Unseen%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Overall Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| BERT Token Classification | 0.9593 | Not recorded | 68fa6e671e209131cea3fc22 |
| Heterogeneous Dynamic Convolutions | 0.9592 | Not recorded | 68fa6e671e209131cea3fc22 |
| ELMo & LSTM Classification | 0.932 | Not recorded | 68fa6e671e209131cea3fc22 |

### BERT Token Classification

Overall Accuracy: 0.9593. Result ID: 699017df3a6c314849fe14ef.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e671e209131cea3fc22.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68f51e209131cea1f541.md>)

Recorded evaluation configuration: {"source":"Numbers from Prange et al. (2021)"}. This record does not assert independent reproduction.

### Heterogeneous Dynamic Convolutions

Overall Accuracy: 0.9592. Result ID: 699017df3a6c314849fe14f0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e671e209131cea3fc22.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e671e209131cea3fc22.md>)

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

### ELMo & LSTM Classification

Overall Accuracy: 0.932. Result ID: 699017df3a6c314849fe14ee.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e671e209131cea3fc22.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.