Update README.md
Browse files
README.md
CHANGED
@@ -30,19 +30,18 @@ For machine translation to tackle discourse phenomena, models must have access t
|
|
30 |
|
31 |
## Uses
|
32 |
|
33 |
-
|
|
|
|
|
|
|
|
|
34 |
|
35 |
-
### Direct Use
|
36 |
-
|
37 |
-
<!-- This section describes suitable use cases for the dataset. -->
|
38 |
-
|
39 |
-
[More Information Needed]
|
40 |
|
41 |
### Out-of-Scope Use
|
42 |
|
43 |
-
|
44 |
-
|
45 |
-
|
46 |
|
47 |
## Dataset Structure
|
48 |
|
@@ -78,15 +77,7 @@ There is no personal or sensitive information in the dataset.
|
|
78 |
|
79 |
[More Information Needed]
|
80 |
|
81 |
-
|
82 |
-
|
83 |
-
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
84 |
-
|
85 |
-
Users should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.
|
86 |
-
|
87 |
-
## Citation [optional]
|
88 |
-
|
89 |
-
<!-- If there is a paper or blog post introducing the dataset, the APA and Bibtex information for that should go in this section. -->
|
90 |
|
91 |
**BibTeX:**
|
92 |
|
@@ -113,4 +104,4 @@ Bawden, R., Sennrich, R., Birch, A., & Haddow, B. (2018). Evaluating Discourse P
|
|
113 |
|
114 |
## Dataset Card Authors
|
115 |
|
116 |
-
Rachel Bawden
|
|
|
30 |
|
31 |
## Uses
|
32 |
|
33 |
+
The dataset is a contrastive evaluation test used to test the use of previous linguistic context (one previous sentence) when translating
|
34 |
+
from English to French. Each example block is composed of a previous source sentence and two possible translations, accompanied with
|
35 |
+
accompanying context, which can be used to disambiguate the two possible translations. Each instance in the dataset has a correct translation
|
36 |
+
and a contrastive one. MT models should be used to rank each translation given the context provided, and the model is evaluated on its
|
37 |
+
accuracy of ranking the correct translations above the contrastive ones.
|
38 |
|
|
|
|
|
|
|
|
|
|
|
39 |
|
40 |
### Out-of-Scope Use
|
41 |
|
42 |
+
The test set is designed for translation from English to French. The test set cannot be used in the opposite direction, as the ambiguity
|
43 |
+
of the source sentence will not hold. The test set is used to be used as a complement to global metrics of evaluation quality and assumes
|
44 |
+
that the model has been designed for translation.
|
45 |
|
46 |
## Dataset Structure
|
47 |
|
|
|
77 |
|
78 |
[More Information Needed]
|
79 |
|
80 |
+
## Citation
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
81 |
|
82 |
**BibTeX:**
|
83 |
|
|
|
104 |
|
105 |
## Dataset Card Authors
|
106 |
|
107 |
+
Rachel Bawden: rachel.bawden@inria.fr
|