Delete HebNLI-README.md
Browse files- HebNLI-README.md +0 -52
HebNLI-README.md
DELETED
|
@@ -1,52 +0,0 @@
|
|
| 1 |
-
# HebNLI - A Natural Language Inference Dataset in Hebrew
|
| 2 |
-
|
| 3 |
-
## Summary
|
| 4 |
-
HebNLI is a Hebrew dataset for natural language inference (NLI) tasks.
|
| 5 |
-
|
| 6 |
-
## Introduction
|
| 7 |
-
This dataset is the first of its kind in the Hebrew language and aims to serve as training data for NLI tasks.
|
| 8 |
-
HebNLI is based on MultiNLI, a large crowd-sourced corpus of sentences from varied genres and writing styles in the English language.
|
| 9 |
-
MultiNLI was originally built by collecting hundreds of thousands of base sentences from which different taggers derived follow-up sentences that stand in one of 3 logical relations to the base sentences: entailment, contradiction or neutral.
|
| 10 |
-
Different taggers were then given paired sentences - base sentence and a derived sentence. The logical relation between them was determined by the majority vote, and each pair of sentences was labled according to the determined logical relation.
|
| 11 |
-
In HebNLI we used machine translation (Google Gemini) to translate the English corpus to Hebrew, such that each base sentence and its compiled derivative sentences appear in Hebrew.
|
| 12 |
-
|
| 13 |
-
## Genres/Sources in HebNLI
|
| 14 |
-
HebNLI comprises 7 of the original 10 genres/sources that appeared in MultiNLI:
|
| 15 |
-
1. Nine eleven - Written protocols from a commitee investigating the events of 9/11.
|
| 16 |
-
2. Government - Reports, speeches and press releases published on U.S.A government websites.
|
| 17 |
-
3. Letters - A database of letters written in the late 90's and early 2000's.
|
| 18 |
-
4. OUP (Oxford University Press) - Publications about the textile industry and about child development.
|
| 19 |
-
5. Slate - Pop-culture articles published in Slate magazine.
|
| 20 |
-
6. Travel - Travel guides by Berlitz press.
|
| 21 |
-
7. Fiction - Texts extracted from modern works of literature.
|
| 22 |
-
|
| 23 |
-
The remaining three sources were found to either be too English-oriented to be properly translated to Hebrew by machine translation ("Verbatim" magazine source), or included too many broken sentences and filler-words to be properly translated to Hebrew by machine translation (face-to-face conversations and telephone conversations sources).
|
| 24 |
-
|
| 25 |
-
## Dataset Statistics
|
| 26 |
-
The table below shows the distribution of each source corpus within HebNLI (how many setences exist in the dataset from each source).
|
| 27 |
-
|
| 28 |
-
| Genre/Source | HebNLI Corpus |
|
| 29 |
-
|------------------|------------------|
|
| 30 |
-
| Nine eleven | 1878 |
|
| 31 |
-
| Government | 76953 |
|
| 32 |
-
| Letters | 1974 |
|
| 33 |
-
| OUP | 1986 |
|
| 34 |
-
| Slate | 71082 |
|
| 35 |
-
| Travel | 75776 |
|
| 36 |
-
| Fiction | 73734 |
|
| 37 |
-
|
| 38 |
-
Total # of sentences = 303,383.
|
| 39 |
-
|
| 40 |
-
### HebNLI Blog Post
|
| 41 |
-
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
|
| 42 |
-
|
| 43 |
-
### Original MultiNLI Paper
|
| 44 |
-
https://cims.nyu.edu/~sbowman/multinli/paper.pdf
|
| 45 |
-
|
| 46 |
-
## Contributors
|
| 47 |
-
HebNLI was translated and checked for quality by Webiks for MAFAT, as part of the National Natural Language Processing Plan of Israel.
|
| 48 |
-
|
| 49 |
-
Contributors: Hilla Merhav Fine (Webiks), Yaniv Maylik (Webiks), Carrine Cherf (Webiks), Tal Geva (MAFAT).
|
| 50 |
-
|
| 51 |
-
## Acknowledgments
|
| 52 |
-
We would like to express our gratitude to Adina Williams, Nikita Nangia and Samuel R. Bowman, the creators of [the original NLI dataset MultiNLI](https://huggingface.co/datasets/nyu-mll/multi_nli).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|