VALSE: A Task-independent benchmark for Vision and Language models centered on linguistic phenomena

L Parcalabescu, M Cafagna, L Muradjan, A Frank, I Calixto, A Gatt

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    Abstract

    We propose VALSE (Vision And Language Structured Evaluation), a novel benchmark designed for testing general-purpose pretrained vision and language (V&L) models for their visio-linguistic grounding capabilities on specific linguistic phenomena. VALSE offers a suite of six tests covering various linguistic constructs. Solving these requires models to ground linguistic phenomena in the visual modality, allowing more fine-grained evaluations than hitherto possible. We build VALSE using methods that support the construction of valid foils, and report results from evaluating five widely-used V&L models. Our experiments suggest that current models have considerable difficulty addressing most phenomena. Hence, we expect VALSE to serve as an important benchmark to measure future progress of pretrained V&L models from a linguistic perspective, complementing the canonical task-centred V&L evaluations.
    Original languageEnglish
    Title of host publicationProceedings of the 60th Annual Meeting of the Association for Computational Linguistics (ACL'22)
    PublisherAssociation for Computational Linguistics
    Pages8253–8280
    DOIs
    Publication statusPublished - 2022

    Fingerprint

    Dive into the research topics of 'VALSE: A Task-independent benchmark for Vision and Language models centered on linguistic phenomena'. Together they form a unique fingerprint.

    Cite this