Dataset Card for IRFL

• Dataset Description
  • Leaderboards
  • Colab notebook code for IRFL evaluation
  • Languages
• Dataset Structure
  • Data Fields
• Dataset Creation
• Considerations for Using the Data
  • Licensing Information
  • Citation Information

Dataset Description

The IRFL dataset consists of idioms, similes, metaphors with matching figurative and literal images, and two novel tasks of multimodal figurative detection and retrieval.

Using human annotation and an automatic pipeline we created, we collected figurative and literal images for textual idioms, metaphors, and similes. We annotated the relations between these images and the figurative phrase they originated from. We created two novel tasks of figurative detection and retrieval using these images.

The figurative detection task evaluates Vision and Language Pre-Trained Models’ (VL-PTMs) ability to choose the image that best visualizes the meaning of a figurative expression. The task is to choose the image that best visualizes the figurative phrase out of X candidates. The retrieval task examines VL-PTMs' preference for figurative images. In this task, Given a set of figurative and partially literal images, the task is to rank the images using the model-matching score such that the figurative images are ranked higher, and calculate the precision at k, where k is the number of figurative images in the input.

We evaluated state-of-the-art VL models and found that the best models achieved 22%, 30%, and 66% accuracy vs. humans 97%, 99.7%, and 100% on our detection task for idioms, metaphors, and similes respectively. The best model achieved an F1 score of 61 on the retrieval task.

• Homepage: https://irfl-dataset.github.io/
• Repository: https://github.com/irfl-dataset/IRFL
• Paper: https://arxiv.org/abs/2303.15445
• Leaderboard: https://irfl-dataset.github.io/leaderboard
• Point of Contact: irfl.dataset@gmail.com; ron.yosef@mail.huji.ac.il

Leaderboards

https://irfl-dataset.github.io/leaderboard

Colab notebook for IRFL evaluation

https://colab.research.google.com/drive/1RfcUhBTHvREx5X7TMY5UAgMYX8NMKy7u?usp=sharing

Languages

English.

Dataset Structure

Data Fields

★ - refers to idiom-only fields ⁺₊ - refers to metaphor-only fields

Multimodal Figurative Language Detection task

• query (★): the idiom definition the answer image originated from.
• distractors: the distractor images
• answer: the correct image
• figurative_type: idiom | metaphor | simile
• type: the correct image type (Figurative or Figurative+Literal).
• definition (★): list of all the definitions of the idiom
• phrase: the figurative phrase.

Multimodal Figurative Language Retrieval task

• type: the rival categories FvsPL (Figurative images vs. Partial Literal) or FLvsPL (Figurative+Literal images vs. Partial Literal)
• figurative_type: idiom | metaphor | simile
• images_metadata: the metadata of the distractors and answer images.
• first_category: the first category images (Figurative images if FvsPL, Figurative Literal images if FLvsPL)
• second_category: the second category images (Partial Literal)
• definition (★): list of all the definitions of the idiom
• theme (⁺₊): the theme of the partial literal distractor, for example, for the metaphor heart of gold, an image of a "gold bar" and an image of a "human heart" will have different theme value
• phrase: the figurative phrase.

The idioms, metaphor, and similes datasets contain all the figurative phrases, annotated images, and corresponding metadata.

Dataset Collection

Using an automatic pipeline we created, we collected figurative and literal images for textual idioms, metaphors, and similes. We annotated the relations between these images and the figurative phrase they originated from.

Annotation process

We paid Amazon Mechanical Turk Workers to annotate the relation between each image and phrase (Figurative vs. Literal).

Considerations for Using the Data

• Idioms: Annotated by five crowdworkers with rigorous qualifications and training.
• Metaphors and Similes: Annotated by three expert team members.
• Detection and Ranking Tasks: Annotated by three crowdworkers not involved in prior IRFL annotations.

Licensing Information

CC-By 4.0

Citation Information

@misc{yosef2023irfl, title={IRFL: Image Recognition of Figurative Language}, author={Ron Yosef and Yonatan Bitton and Dafna Shahaf}, year={2023}, eprint={2303.15445}, archivePrefix={arXiv}, primaryClass={cs.CL} }