# Label Projection on MLQA (Multi-lingual Metrics)

Source: [Label Projection on MLQA (Multi-lingual Metrics)](<https://www.sota2.com/research/sota/label-projection-on-mlqa-multi-lingual-metrics>)

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

Data updated: 2026-03-13T09:16:02.399Z

Evaluation of the model's ability to project labels across multiple languages (Arabic, Chinese, German, Hindi, Spanish, Vietnamese) on the MLQA dataset, using COMET Score and Label Matches F1 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 | Label Projection |
| Dataset (including any stated split) | MLQA |
| Metric | Average COMET Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 32 of 32 |
| 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 COMET Score](<https://www.sota2.com/research/sota/label-projection-on-mlqa-multi-lingual-metrics.md?metric=Average%20COMET%20Score>): Higher is better; unit: PERCENT.

- [Arabic COMET Score](<https://www.sota2.com/research/sota/label-projection-on-mlqa-multi-lingual-metrics.md?metric=Arabic%20COMET%20Score>): Higher is better; unit: PERCENT.

- [Arabic Label Matches F1](<https://www.sota2.com/research/sota/label-projection-on-mlqa-multi-lingual-metrics.md?metric=Arabic%20Label%20Matches%20F1>): Higher is better; unit: PERCENT.

- [Chinese COMET Score](<https://www.sota2.com/research/sota/label-projection-on-mlqa-multi-lingual-metrics.md?metric=Chinese%20COMET%20Score>): Higher is better; unit: PERCENT.

- [Chinese Label Matches F1](<https://www.sota2.com/research/sota/label-projection-on-mlqa-multi-lingual-metrics.md?metric=Chinese%20Label%20Matches%20F1>): Higher is better; unit: PERCENT.

- [German COMET Score](<https://www.sota2.com/research/sota/label-projection-on-mlqa-multi-lingual-metrics.md?metric=German%20COMET%20Score>): Higher is better; unit: PERCENT.

- [German Label Matches F1](<https://www.sota2.com/research/sota/label-projection-on-mlqa-multi-lingual-metrics.md?metric=German%20Label%20Matches%20F1>): Higher is better; unit: PERCENT.

- [Hindi COMET Score](<https://www.sota2.com/research/sota/label-projection-on-mlqa-multi-lingual-metrics.md?metric=Hindi%20COMET%20Score>): Higher is better; unit: PERCENT.

- [Hindi Label Matches F1](<https://www.sota2.com/research/sota/label-projection-on-mlqa-multi-lingual-metrics.md?metric=Hindi%20Label%20Matches%20F1>): Higher is better; unit: PERCENT.

- [Spanish COMET Score](<https://www.sota2.com/research/sota/label-projection-on-mlqa-multi-lingual-metrics.md?metric=Spanish%20COMET%20Score>): Higher is better; unit: PERCENT.

- [Spanish Label Matches F1](<https://www.sota2.com/research/sota/label-projection-on-mlqa-multi-lingual-metrics.md?metric=Spanish%20Label%20Matches%20F1>): Higher is better; unit: PERCENT.

- [Vietnamese COMET Score](<https://www.sota2.com/research/sota/label-projection-on-mlqa-multi-lingual-metrics.md?metric=Vietnamese%20COMET%20Score>): Higher is better; unit: PERCENT.

- [Vietnamese Label Matches F1](<https://www.sota2.com/research/sota/label-projection-on-mlqa-multi-lingual-metrics.md?metric=Vietnamese%20Label%20Matches%20F1>): Higher is better; unit: PERCENT.

- [Average Label Matches F1](<https://www.sota2.com/research/sota/label-projection-on-mlqa-multi-lingual-metrics.md?metric=Average%20Label%20Matches%20F1>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Average COMET Score | Result date | Report paper |
| --- | --- | --- | --- |
| Awesome-align | 84.8 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| LabelPigeon | 84.8 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| LabelPigeon | 84.7 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| LabelPigeon | 84.3 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| LabelPigeon | 84 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| LabelPigeon | 84 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| EasyProject | 83.7 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| Awesome-align | 83 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| EasyProject | 82.7 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| Awesome-align | 82.5 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| Awesome-align | 82.4 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| EasyProject | 82.3 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| LabelPigeon | 82.3 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| Awesome-align | 82.2 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| EasyProject | 81.7 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| Awesome-align | 81.6 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| EasyProject | 81.4 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| EasyProject | 80.8 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| Awesome-align | 80.4 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| LabelPigeon | 79.6 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| Gemma 3 27B | 79.3 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| EasyProject | 79.1 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| LabelPigeon | 76.9 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| Awesome-align | 76.7 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| EasyProject | 75.7 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| Gemma 3 27B | 62.2 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| Gemma 3 27B | 59.9 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| Gemma 3 27B | 59.1 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| Gemma 3 27B | 54.6 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| Gemma 3 27B | 53.7 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| Gemma 3 27B | 47.2 | Not recorded | 69b39a2c0636e5bf4afbcc0b |
| Gemma 3 27B | 45.3 | Not recorded | 69b39a2c0636e5bf4afbcc0b |

### Awesome-align

Average COMET Score: 84.8. Result ID: 69b3d5c047f34ce9249e352b.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a911e209131cea28ec6.md>)

Recorded evaluation configuration: {"Language":"Arabic"}. This record does not assert independent reproduction.

### LabelPigeon

Average COMET Score: 84.8. Result ID: 69b3d5c047f34ce9249e352e.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

Recorded evaluation configuration: {"Language":"Arabic"}. This record does not assert independent reproduction.

### LabelPigeon

Average COMET Score: 84.7. Result ID: 69b3d5c047f34ce9249e3542.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

Recorded evaluation configuration: {"Language":"Vietnamese"}. This record does not assert independent reproduction.

### LabelPigeon

Average COMET Score: 84.3. Result ID: 69b3d5c047f34ce9249e34fa.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

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

### LabelPigeon

Average COMET Score: 84. Result ID: 69b3d5c047f34ce9249e3536.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

Recorded evaluation configuration: {"Language":"German"}. This record does not assert independent reproduction.

### LabelPigeon

Average COMET Score: 84. Result ID: 69b3d5c047f34ce9249e353e.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

Recorded evaluation configuration: {"Language":"Spanish"}. This record does not assert independent reproduction.

### EasyProject

Average COMET Score: 83.7. Result ID: 69b3d5c047f34ce9249e352d.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756d1e209131cea67d4a.md>)

[Implementation](<https://github.com/edchengg/easyproject>)

Recorded evaluation configuration: {"Language":"Arabic"}. This record does not assert independent reproduction.

### Awesome-align

Average COMET Score: 83. Result ID: 69b3d5c047f34ce9249e353f.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a911e209131cea28ec6.md>)

Recorded evaluation configuration: {"Language":"Vietnamese"}. This record does not assert independent reproduction.

### EasyProject

Average COMET Score: 82.7. Result ID: 69b3d5c047f34ce9249e3541.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756d1e209131cea67d4a.md>)

[Implementation](<https://github.com/edchengg/easyproject>)

Recorded evaluation configuration: {"Language":"Vietnamese"}. This record does not assert independent reproduction.

### Awesome-align

Average COMET Score: 82.5. Result ID: 69b3d5c047f34ce9249e353b.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a911e209131cea28ec6.md>)

Recorded evaluation configuration: {"Language":"Spanish"}. This record does not assert independent reproduction.

### Awesome-align

Average COMET Score: 82.4. Result ID: 69b3d5c047f34ce9249e3533.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a911e209131cea28ec6.md>)

Recorded evaluation configuration: {"Language":"German"}. This record does not assert independent reproduction.

### EasyProject

Average COMET Score: 82.3. Result ID: 69b3d5c047f34ce9249e353d.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756d1e209131cea67d4a.md>)

[Implementation](<https://github.com/edchengg/easyproject>)

Recorded evaluation configuration: {"Language":"Spanish"}. This record does not assert independent reproduction.

### LabelPigeon

Average COMET Score: 82.3. Result ID: 69b3d5c047f34ce9249e3546.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

Recorded evaluation configuration: {"Language":"Average"}. This record does not assert independent reproduction.

### Awesome-align

Average COMET Score: 82.2. Result ID: 69b3d5c047f34ce9249e34f7.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a911e209131cea28ec6.md>)

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

### EasyProject

Average COMET Score: 81.7. Result ID: 69b3d5c047f34ce9249e34f9.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756d1e209131cea67d4a.md>)

[Implementation](<https://github.com/edchengg/easyproject>)

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

### Awesome-align

Average COMET Score: 81.6. Result ID: 69b3d5c047f34ce9249e3543.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a911e209131cea28ec6.md>)

Recorded evaluation configuration: {"Language":"Average"}. This record does not assert independent reproduction.

### EasyProject

Average COMET Score: 81.4. Result ID: 69b3d5c047f34ce9249e3535.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756d1e209131cea67d4a.md>)

[Implementation](<https://github.com/edchengg/easyproject>)

Recorded evaluation configuration: {"Language":"German"}. This record does not assert independent reproduction.

### EasyProject

Average COMET Score: 80.8. Result ID: 69b3d5c047f34ce9249e3545.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756d1e209131cea67d4a.md>)

[Implementation](<https://github.com/edchengg/easyproject>)

Recorded evaluation configuration: {"Language":"Average"}. This record does not assert independent reproduction.

### Awesome-align

Average COMET Score: 80.4. Result ID: 69b3d5c047f34ce9249e352f.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a911e209131cea28ec6.md>)

Recorded evaluation configuration: {"Language":"Chinese"}. This record does not assert independent reproduction.

### LabelPigeon

Average COMET Score: 79.6. Result ID: 69b3d5c047f34ce9249e3532.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

Recorded evaluation configuration: {"Language":"Chinese"}. This record does not assert independent reproduction.

### Gemma 3 27B

Average COMET Score: 79.3. Result ID: 69b3d5c047f34ce9249e34f8.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

Recorded evaluation configuration: {"Backbone":"Gemma 3","Number of parameters":"27B"}. This record does not assert independent reproduction.

### EasyProject

Average COMET Score: 79.1. Result ID: 69b3d5c047f34ce9249e3531.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756d1e209131cea67d4a.md>)

[Implementation](<https://github.com/edchengg/easyproject>)

Recorded evaluation configuration: {"Language":"Chinese"}. This record does not assert independent reproduction.

### LabelPigeon

Average COMET Score: 76.9. Result ID: 69b3d5c047f34ce9249e353a.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

Recorded evaluation configuration: {"Language":"Hindi"}. This record does not assert independent reproduction.

### Awesome-align

Average COMET Score: 76.7. Result ID: 69b3d5c047f34ce9249e3537.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a911e209131cea28ec6.md>)

Recorded evaluation configuration: {"Language":"Hindi"}. This record does not assert independent reproduction.

### EasyProject

Average COMET Score: 75.7. Result ID: 69b3d5c047f34ce9249e3539.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756d1e209131cea67d4a.md>)

[Implementation](<https://github.com/edchengg/easyproject>)

Recorded evaluation configuration: {"Language":"Hindi"}. This record does not assert independent reproduction.

### Gemma 3 27B

Average COMET Score: 62.2. Result ID: 69b3d5c047f34ce9249e3540.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

Recorded evaluation configuration: {"Language":"Vietnamese"}. This record does not assert independent reproduction.

### Gemma 3 27B

Average COMET Score: 59.9. Result ID: 69b3d5c047f34ce9249e3534.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

Recorded evaluation configuration: {"Language":"German"}. This record does not assert independent reproduction.

### Gemma 3 27B

Average COMET Score: 59.1. Result ID: 69b3d5c047f34ce9249e353c.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

Recorded evaluation configuration: {"Language":"Spanish"}. This record does not assert independent reproduction.

### Gemma 3 27B

Average COMET Score: 54.6. Result ID: 69b3d5c047f34ce9249e3544.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

Recorded evaluation configuration: {"Language":"Average"}. This record does not assert independent reproduction.

### Gemma 3 27B

Average COMET Score: 53.7. Result ID: 69b3d5c047f34ce9249e3530.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

Recorded evaluation configuration: {"Language":"Chinese"}. This record does not assert independent reproduction.

### Gemma 3 27B

Average COMET Score: 47.2. Result ID: 69b3d5c047f34ce9249e352c.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

Recorded evaluation configuration: {"Language":"Arabic"}. This record does not assert independent reproduction.

### Gemma 3 27B

Average COMET Score: 45.3. Result ID: 69b3d5c047f34ce9249e3538.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a2c0636e5bf4afbcc0b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

Recorded evaluation configuration: {"Language":"Hindi"}. This record does not assert independent reproduction.

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