Datasets:
Tasks:
Sentence Similarity
Modalities:
Text
Formats:
csv
Languages:
Romanian
Size:
10K - 100K
License:
Update README.md
Browse files
README.md
CHANGED
@@ -9,8 +9,6 @@ task_categories:
|
|
9 |
pretty_name: ROSTSC
|
10 |
---
|
11 |
|
12 |
-
|
13 |
-
|
14 |
# RO-STS-Corpus
|
15 |
|
16 |
## Overview
|
@@ -21,6 +19,15 @@ RO-STS-Corpus is a Romanian Semantic Textual Similarity (STS) dataset designed f
|
|
21 |
- **sentence2**: The second sentence in the pair.
|
22 |
- **score**: A numerical value representing the semantic similarity between the two sentences, normalized between 0 and 1.
|
23 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
24 |
## Usage
|
25 |
This dataset can be used for:
|
26 |
- Fine-tuning and evaluating sentence embedding models for Romanian.
|
@@ -44,4 +51,5 @@ If you use RO-STS-Corpus in your research, please cite this dataset as follows:
|
|
44 |
We acknowledge contributions from various resources used in compiling and annotating this dataset.
|
45 |
|
46 |
---
|
47 |
-
Further details and additional metadata will be added in future versions.
|
|
|
|
9 |
pretty_name: ROSTSC
|
10 |
---
|
11 |
|
|
|
|
|
12 |
# RO-STS-Corpus
|
13 |
|
14 |
## Overview
|
|
|
19 |
- **sentence2**: The second sentence in the pair.
|
20 |
- **score**: A numerical value representing the semantic similarity between the two sentences, normalized between 0 and 1.
|
21 |
|
22 |
+
## Dataset Composition
|
23 |
+
| Source Dataset | Split | Number of Pairs |
|
24 |
+
|---------------|-------|----------------|
|
25 |
+
| mteb/sts17-crosslingual-sts | en-en | 250 |
|
26 |
+
| mteb/sts12-sts | train | 2225 |
|
27 |
+
| mteb/sts12-sts | test | 3029 |
|
28 |
+
| mteb/biosses-sts | test | 100 |
|
29 |
+
| mteb/sts16-sts | test | 1184 |
|
30 |
+
|
31 |
## Usage
|
32 |
This dataset can be used for:
|
33 |
- Fine-tuning and evaluating sentence embedding models for Romanian.
|
|
|
51 |
We acknowledge contributions from various resources used in compiling and annotating this dataset.
|
52 |
|
53 |
---
|
54 |
+
Further details and additional metadata will be added in future versions.
|
55 |
+
|