QLEVR: A Diagnostic Dataset for Quantificational Language and Elementary Visual Reasoning

Findings (NAACL) 2022  ·  Zechen Li, Anders Søgaard ·

Synthetic datasets have successfully been used to probe visual question-answering datasets for their reasoning abilities. CLEVR (johnson2017clevr), for example, tests a range of visual reasoning abilities. The questions in CLEVR focus on comparisons of shapes, colors, and sizes, numerical reasoning, and existence claims. This paper introduces a minimally biased, diagnostic visual question-answering dataset, QLEVR, that goes beyond existential and numerical quantification and focus on more complex quantifiers and their combinations, e.g., asking whether there are more than two red balls that are smaller than at least three blue balls in an image. We describe how the dataset was created and present a first evaluation of state-of-the-art visual question-answering models, showing that QLEVR presents a formidable challenge to our current models. Code and Dataset are available at https://github.com/zechenli03/QLEVR

PDF Abstract Findings (NAACL) 2022 PDF Findings (NAACL) 2022 Abstract

Datasets


Introduced in the Paper:

QLEVR

Used in the Paper:

Visual Question Answering CLEVR SHAPES
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Visual Question Answering (VQA) QLEVR MAC Overall Accuracy 66.5 # 1
Visual Question Answering (VQA) QLEVR CNN+LSTM Overall Accuracy 65.9 # 2
Visual Question Answering (VQA) QLEVR BERT Overall Accuracy 65.8 # 3
Visual Question Answering (VQA) QLEVR LSTM Overall Accuracy 64.6 # 4
Visual Question Answering (VQA) QLEVR Q-type Overall Accuracy 50.0 # 5

Methods


No methods listed for this paper. Add relevant methods here