Dataset Card for "COMPS"

Dataset Description

COMPS is a dataset of minimal pair sentences in English that enables the testing knowledge of concepts and their properties in language models (LMs). Specifically, it tests the ability of LMs to attribute properties to everyday concepts, and demonstrate reasoning compatible with property inheritance, where subordinate concepts inherit the properties of their superordinate (hypernyms).

• Homepage: https://github.com/kanishkamisra/comps/
• Repository: https://github.com/kanishkamisra/comps/
• Paper: arxiv
• Point of Contact: [Kanishka Misra] (https://kanishka.website)

Citation Information

@inproceedings{misra-etal-2023-comps,
    title = "{COMPS}: Conceptual Minimal Pair Sentences for testing Robust Property Knowledge and its Inheritance in Pre-trained Language Models",
    author = "Misra, Kanishka  and
      Rayz, Julia  and
      Ettinger, Allyson",
    booktitle = "Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics",
    month = may,
    year = "2023",
    address = "Dubrovnik, Croatia",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2023.eacl-main.213",
    doi = "10.18653/v1/2023.eacl-main.213",
    pages = "2928--2949",
    abstract = "A characteristic feature of human semantic cognition is its ability to not only store and retrieve the properties of concepts observed through experience, but to also facilitate the inheritance of properties (can breathe) from superordinate concepts (animal) to their subordinates (dog){---}i.e. demonstrate property inheritance. In this paper, we present COMPS, a collection of minimal pair sentences that jointly tests pre-trained language models (PLMs) on their ability to attribute properties to concepts and their ability to demonstrate property inheritance behavior. Analyses of 22 different PLMs on COMPS reveal that they can easily distinguish between concepts on the basis of a property when they are trivially different, but find it relatively difficult when concepts are related on the basis of nuanced knowledge representations. Furthermore, we find that PLMs can show behaviors suggesting successful property inheritance in simple contexts, but fail in the presence of distracting information, which decreases the performance of many models sometimes even below chance. This lack of robustness in demonstrating simple reasoning raises important questions about PLMs{'} capacity to make correct inferences even when they appear to possess the prerequisite knowledge.",
}