README.md

metadata

annotations_creators:
  - other
language_creators:
  - machine-generated
language:
  - en
license:
  - cc-by-sa-4.0
multilinguality:
  - monolingual
pretty_name: Adversarial GLUE
size_categories:
  - n<1K
source_datasets:
  - extended|glue
task_categories:
  - text-classification
task_ids:
  - natural-language-inference
  - sentiment-classification
  - text-classification-other-paraphrase-identification
  - text-classification-other-qa-nli
configs:
  - adv_mnli
  - adv_mnli_mismatched
  - adv_qnli
  - adv_qqp
  - adv_rte
  - adv_sst2

Dataset Card for Adversarial GLUE

Table of Contents

• Table of Contents
• Dataset Description
  • Dataset Summary
  • Supported Tasks and Leaderboards
  • Languages
• Dataset Structure
  • Data Instances
  • Data Fields
  • Data Splits
• Dataset Creation
  • Curation Rationale
  • Source Data
  • Annotations
  • Personal and Sensitive Information
• Considerations for Using the Data
  • Social Impact of Dataset
  • Discussion of Biases
  • Other Known Limitations
• Additional Information
  • Dataset Curators
  • Licensing Information
  • Citation Information
  • Contributions

Dataset Description

• Homepage: https://adversarialglue.github.io/
• Repository:
• Paper: arXiv
• Leaderboard:
• Point of Contact:

Dataset Summary

Adversarial GLUE Benchmark (AdvGLUE) is a comprehensive robustness evaluation benchmark that focuses on the adversarial robustness evaluation of language models. It covers five natural language understanding tasks from the famous GLUE tasks and is an adversarial version of GLUE benchmark.

AdvGLUE considers textual adversarial attacks from different perspectives and hierarchies, including word-level transformations, sentence-level manipulations, and human-written adversarial examples, which provide comprehensive coverage of various adversarial linguistic phenomena.

Supported Tasks and Leaderboards

Leaderboard available on the homepage: https://adversarialglue.github.io/.

Languages

AdvGLUE deviates from the GLUE dataset, which has a base language of English.

Dataset Structure

Data Instances

default

• Size of downloaded dataset files: 198 KB
• Example:

>>> datasets.load_dataset('adv_glue', 'adv_sst2')['validation'][0]
{'sentence': "it 's an uneven treat that bores fun at the democratic exercise while also examining its significance for those who take part .", 'label': 1, 'idx': 0}

Data Fields

The data fields are the same as in the GLUE dataset, which differ by task.

The data fields are the same among all splits.

adv_mnli

• premise: a string feature.
• hypothesis: a string feature.
• label: a classification label, with possible values including entailment (0), neutral (1), contradiction (2).
• idx: a int32 feature.

adv_mnli_matched

• premise: a string feature.
• hypothesis: a string feature.
• label: a classification label, with possible values including entailment (0), neutral (1), contradiction (2).
• idx: a int32 feature.

adv_mnli_mismatched

• premise: a string feature.
• hypothesis: a string feature.
• label: a classification label, with possible values including entailment (0), neutral (1), contradiction (2).
• idx: a int32 feature.

adv_qnli

More Information Needed

adv_qqp

More Information Needed

adv_rte

More Information Needed

adv_sst2

More Information Needed

Data Splits

Adversarial GLUE provides only a 'dev' split.

Dataset Creation

Curation Rationale

[More Information Needed]

Source Data

Initial Data Collection and Normalization

[More Information Needed]

Who are the source language producers?

[More Information Needed]

Annotations

Annotation process

[More Information Needed]

Who are the annotators?

[More Information Needed]

Personal and Sensitive Information

[More Information Needed]

Considerations for Using the Data

Social Impact of Dataset

[More Information Needed]

Discussion of Biases

[More Information Needed]

Other Known Limitations

[More Information Needed]

Additional Information

Dataset Curators

[More Information Needed]

Licensing Information

The dataset is distributed under the CC BY-SA 4.0 license.

Citation Information

@article{Wang2021AdversarialGA,
  title={Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models},
  author={Boxin Wang and Chejian Xu and Shuohang Wang and Zhe Gan and Yu Cheng and Jianfeng Gao and Ahmed Hassan Awadallah and B. Li},
  journal={ArXiv},
  year={2021},
  volume={abs/2111.02840}
}

Contributions

Thanks to @jxmorris12 for adding this dataset.