Add link to code repository (#2)
Browse files- Add link to code repository (7b298de7bcd0e2533d4a9f86c5b02719ff259cae)
Co-authored-by: Niels Rogge <[email protected]>
README.md
CHANGED
|
@@ -1,66 +1,66 @@
|
|
| 1 |
---
|
| 2 |
-
language:
|
| 3 |
-
- en
|
| 4 |
-
- ar
|
| 5 |
-
- es
|
| 6 |
-
- fr
|
| 7 |
-
- ru
|
| 8 |
-
- hi
|
| 9 |
-
- ms
|
| 10 |
-
- sw
|
| 11 |
-
- az
|
| 12 |
-
- ko
|
| 13 |
-
- pt
|
| 14 |
-
- hy
|
| 15 |
-
- th
|
| 16 |
-
- uk
|
| 17 |
-
- ur
|
| 18 |
-
- sr
|
| 19 |
-
- iw
|
| 20 |
-
- ja
|
| 21 |
-
- hr
|
| 22 |
-
- tl
|
| 23 |
-
- ky
|
| 24 |
-
- vi
|
| 25 |
-
- fa
|
| 26 |
-
- tg
|
| 27 |
-
- mg
|
| 28 |
-
- nl
|
| 29 |
-
- ne
|
| 30 |
-
- uz
|
| 31 |
-
- my
|
| 32 |
-
- da
|
| 33 |
-
- dz
|
| 34 |
-
- id
|
| 35 |
-
- is
|
| 36 |
-
- tr
|
| 37 |
-
- lo
|
| 38 |
-
- sl
|
| 39 |
-
- so
|
| 40 |
-
- mn
|
| 41 |
-
- bn
|
| 42 |
-
- bs
|
| 43 |
-
- ht
|
| 44 |
-
- el
|
| 45 |
-
- it
|
| 46 |
-
- to
|
| 47 |
-
- ka
|
| 48 |
-
- sn
|
| 49 |
-
- sq
|
| 50 |
-
- zh
|
| 51 |
-
pretty_name: BordIRlines
|
| 52 |
-
multilinguality:
|
| 53 |
-
- multilingual
|
| 54 |
annotations_creators:
|
| 55 |
-
|
| 56 |
-
|
| 57 |
language_creators:
|
| 58 |
-
|
| 59 |
-
|
| 60 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 61 |
license: mit
|
|
|
|
|
|
|
|
|
|
|
|
|
| 62 |
task_categories:
|
| 63 |
-
|
|
|
|
| 64 |
arxiv: 2410.01171
|
| 65 |
---
|
| 66 |
|
|
@@ -68,6 +68,8 @@ arxiv: 2410.01171
|
|
| 68 |
|
| 69 |
This is the dataset associated with the paper "BordIRlines: A Dataset for Evaluating Cross-lingual Retrieval-Augmented Generation" ([link](https://arxiv.org/abs/2410.01171)).
|
| 70 |
|
|
|
|
|
|
|
| 71 |
## Dataset Summary
|
| 72 |
|
| 73 |
The **BordIRLines Dataset** is an information retrieval (IR) dataset constructed from various language corpora. It contains queries and corresponding ranked docs along with their relevance scores. The dataset includes multiple languages, including English, Arabic, Spanish, and others, and is split across different sources like LLM-based outputs.
|
|
@@ -126,6 +128,7 @@ The dataset includes docs and queries in the following **languages**:
|
|
| 126 |
- `ka`: Georgian
|
| 127 |
- `sn`: Shona
|
| 128 |
- `sq`: Albanian
|
|
|
|
| 129 |
- `control`: see below
|
| 130 |
|
| 131 |
The **control** language is English, and contains the queries for all 251 territories. In contrast, **en** is only the 38 territories which have an English-speaking claimant.
|
|
@@ -241,4 +244,4 @@ ds_llm_fewshot_zhs = load_dataset("borderlines/bordirlines", "zhs", relevance_fi
|
|
| 241 |
primaryClass={cs.CL},
|
| 242 |
url={https://arxiv.org/abs/2410.01171},
|
| 243 |
}
|
| 244 |
-
```
|
|
|
|
| 1 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
annotations_creators:
|
| 3 |
+
- human
|
| 4 |
+
- machine-generated
|
| 5 |
language_creators:
|
| 6 |
+
- found
|
| 7 |
+
language:
|
| 8 |
+
- en
|
| 9 |
+
- ar
|
| 10 |
+
- es
|
| 11 |
+
- fr
|
| 12 |
+
- ru
|
| 13 |
+
- hi
|
| 14 |
+
- ms
|
| 15 |
+
- sw
|
| 16 |
+
- az
|
| 17 |
+
- ko
|
| 18 |
+
- pt
|
| 19 |
+
- hy
|
| 20 |
+
- th
|
| 21 |
+
- uk
|
| 22 |
+
- ur
|
| 23 |
+
- sr
|
| 24 |
+
- iw
|
| 25 |
+
- ja
|
| 26 |
+
- hr
|
| 27 |
+
- tl
|
| 28 |
+
- ky
|
| 29 |
+
- vi
|
| 30 |
+
- fa
|
| 31 |
+
- tg
|
| 32 |
+
- mg
|
| 33 |
+
- nl
|
| 34 |
+
- ne
|
| 35 |
+
- uz
|
| 36 |
+
- my
|
| 37 |
+
- da
|
| 38 |
+
- dz
|
| 39 |
+
- id
|
| 40 |
+
- is
|
| 41 |
+
- tr
|
| 42 |
+
- lo
|
| 43 |
+
- sl
|
| 44 |
+
- so
|
| 45 |
+
- mn
|
| 46 |
+
- bn
|
| 47 |
+
- bs
|
| 48 |
+
- ht
|
| 49 |
+
- el
|
| 50 |
+
- it
|
| 51 |
+
- to
|
| 52 |
+
- ka
|
| 53 |
+
- sn
|
| 54 |
+
- sq
|
| 55 |
+
- zh
|
| 56 |
license: mit
|
| 57 |
+
multilinguality:
|
| 58 |
+
- multilingual
|
| 59 |
+
source_datasets:
|
| 60 |
+
- manestay/borderlines
|
| 61 |
task_categories:
|
| 62 |
+
- question-answering
|
| 63 |
+
pretty_name: BordIRlines
|
| 64 |
arxiv: 2410.01171
|
| 65 |
---
|
| 66 |
|
|
|
|
| 68 |
|
| 69 |
This is the dataset associated with the paper "BordIRlines: A Dataset for Evaluating Cross-lingual Retrieval-Augmented Generation" ([link](https://arxiv.org/abs/2410.01171)).
|
| 70 |
|
| 71 |
+
Code: https://github.com/manestay/bordIRlines
|
| 72 |
+
|
| 73 |
## Dataset Summary
|
| 74 |
|
| 75 |
The **BordIRLines Dataset** is an information retrieval (IR) dataset constructed from various language corpora. It contains queries and corresponding ranked docs along with their relevance scores. The dataset includes multiple languages, including English, Arabic, Spanish, and others, and is split across different sources like LLM-based outputs.
|
|
|
|
| 128 |
- `ka`: Georgian
|
| 129 |
- `sn`: Shona
|
| 130 |
- `sq`: Albanian
|
| 131 |
+
- `zh`: Chinese
|
| 132 |
- `control`: see below
|
| 133 |
|
| 134 |
The **control** language is English, and contains the queries for all 251 territories. In contrast, **en** is only the 38 territories which have an English-speaking claimant.
|
|
|
|
| 244 |
primaryClass={cs.CL},
|
| 245 |
url={https://arxiv.org/abs/2410.01171},
|
| 246 |
}
|
| 247 |
+
```
|