# Cross-lingual Question Answering on Open-Ended CLQA

Source: [Cross-lingual Question Answering on Open-Ended CLQA](<https://www.sota2.com/research/sota/cross-lingual-question-answering-on-open-ended-clqa>)

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

Data updated: 2026-04-14T09:41:14.130Z

Evaluation of cross-lingual question answering performance on the Open-Ended CLQA dataset, reporting rank counts for retrieved answers.

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 | Cross-lingual Question Answering |
| Dataset (including any stated split) | Open-Ended CLQA |
| Metric | Rank 1 Count |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Rank 1 Count](<https://www.sota2.com/research/sota/cross-lingual-question-answering-on-open-ended-clqa.md?metric=Rank%201%20Count>): Higher is better; unit: SCALAR.

- [Rank 2 Count](<https://www.sota2.com/research/sota/cross-lingual-question-answering-on-open-ended-clqa.md?metric=Rank%202%20Count>): Higher is better; unit: SCALAR.

- [Rank 3 Count](<https://www.sota2.com/research/sota/cross-lingual-question-answering-on-open-ended-clqa.md?metric=Rank%203%20Count>): Higher is better; unit: SCALAR.

- [Rank 4 Count](<https://www.sota2.com/research/sota/cross-lingual-question-answering-on-open-ended-clqa.md?metric=Rank%204%20Count>): Higher is better; unit: SCALAR.

- [Same Count](<https://www.sota2.com/research/sota/cross-lingual-question-answering-on-open-ended-clqa.md?metric=Same%20Count>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Rank 1 Count | Result date | Report paper |
| --- | --- | --- | --- |
| Llama\_CL | 325 | Not recorded | 69ddca167d0172140ec79a0c |
| L\_NTP | 271 | Not recorded | 69ddca167d0172140ec79a0c |
| L\_Bi\_NTP | 252 | Not recorded | 69ddca167d0172140ec79a0c |
| L-3-8B | 181 | Not recorded | 69ddca167d0172140ec79a0c |

### Llama\_CL

Rank 1 Count: 325. Result ID: 69de0b5247f34ce924c5d52d.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca167d0172140ec79a0c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ddca167d0172140ec79a0c.md>)

Recorded evaluation configuration: {"Evaluation protocol":"1-4 scale rating","Evaluation judge":"Claude 3.5","Total samples":"1,000"}. This record does not assert independent reproduction.

### L\_NTP

Rank 1 Count: 271. Result ID: 69de0b5247f34ce924c5d52b.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca167d0172140ec79a0c.md>)

Recorded evaluation configuration: {"Evaluation protocol":"1-4 scale rating","Evaluation judge":"Claude 3.5","Total samples":"1,000"}. This record does not assert independent reproduction.

### L\_Bi\_NTP

Rank 1 Count: 252. Result ID: 69de0b5247f34ce924c5d52c.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca167d0172140ec79a0c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73fe1e209131cea5f169.md>)

Recorded evaluation configuration: {"Evaluation protocol":"1-4 scale rating","Evaluation judge":"Claude 3.5","Total samples":"1,000"}. This record does not assert independent reproduction.

### L-3-8B

Rank 1 Count: 181. Result ID: 69de0b5247f34ce924c5d52a.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca167d0172140ec79a0c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Evaluation protocol":"1-4 scale rating","Evaluation judge":"Claude 3.5","Total samples":"1,000"}. This record does not assert independent reproduction.

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