VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena

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.

Paper accepted for publication at ACL 2022 Main; 28 pages, 4 figures, 11 tables

References in corpus (3)

• Microsoft COCO Captions: Data Collection and Evaluation Server
• Visual Entailment: A Novel Task for Fine-Grained Image Understanding
• VALUE: A Multi-Task Benchmark for Video-and-Language Understanding Evaluation

Cited by in corpus (6)

• MM-SHAP: A Performance-agnostic Metric for Measuring Multimodal Contributions in Vision and Language Models & Tasks
• Interpreting Vision and Language Generative Models with Semantic Visual Priors
• BOK-VQA: Bilingual outside Knowledge-Based Visual Question Answering via Graph Representation Pretraining
• Explicitly Representing Syntax Improves Sentence-to-layout Prediction of Unexpected Situations
• Natural Language Processing RELIES on Linguistics
• HNC: Leveraging Hard Negative Captions towards Models with Fine-Grained Visual-Linguistic Comprehension Capabilities