Hugging Face's logo

Datasets:
cardiffnlp
/
relentless

Modalities:
Text
Time-series
Languages:
English
Size:
< 1K
ArXiv:
Libraries:
Datasets
License:
Dataset card Viewer Files Community
1
relentless / README.md
asahi417's picture
asahi417
Update README.md
16b57b0
preview code
|
raw
history blame
2.47 kB
metadata
language:
  - en
license:
  - other
multilinguality:
  - monolingual
size_categories:
  - n<1K
pretty_name: TBA

Dataset Card for "cardiffnlp/relentless"

RelEntLess is a new benchmark, in which entity pairs have to be ranked according to how much they satisfy a given graded relation. Essentially, the task is a ranking task where we provide five prototypical examples to each relation. Following brief description of each relation type is used in our baseline in addition to the prototypical examples. Please check our paper "A RelEntLess Benchmark for Modelling Graded Relations between Named Entities" for more detail. THe test set is private at the moment.

{
    "is friend/ally of": "entities that are friends or allies",
    "is competitor/rival of": "entities that are competitors or rivals",
    "is known for": "examples of what entities are known for",
    "is influenced by": "what has influenced different entities",
    "is similar to": "examples of entities that are similar"
}

Dataset Description

Dataset Summary

relation_type val.
is friend/ally of 20
is competitor/rival of 20
is known for 20
is influenced by 20
is similar to 20

Dataset Structure

Data Instances 

{
    "pairs": [["Le Corbusier", "purism art"], ["Sean Connery", "Finding Forrester"], ...],
    "scores_all": [[4.0, 5.0, 3.0, 4.0, 5.0, 3.0, 5.0], [4.0, 5.0, 2, 5.0, 5.0, 4.0, 2], ...],
    "scores_mean": [4.142857142857143, 3.857142857142857, 4.857142857142857, ...],
    "relation_type": "is known for",
    "ranks": [8.5, 11, 5, 14, 15, 5, 20, 13, 1.5, 18, 10, 1.5, 17, ...],
    "prototypical_examples": [["Extinction Rebellion", "Greta Thunberg"], ["Israel", "Palestine"],...]
}

Citation Information 

TBA

LICENSE

The LICENSE of all the resources are under CC-BY-NC-4.0. Thus, they are freely available for academic purpose or individual research, but restricted for commercial use.