Generalized Hadamard-Product Fusion Operators for Visual Question Answering
This work addresses the challenge of effectively combining visual and textual information for VQA, presenting an incremental advancement through novel operator components.
The authors tackled the problem of multimodal fusion for visual question answering by proposing a generalized class of Hadamard-product-based fusion operators, resulting in a 1.1% absolute improvement in OpenEnded accuracy on the VQA 2.0 test-dev set over baseline methods.
We propose a generalized class of multimodal fusion operators for the task of visual question answering (VQA). We identify generalizations of existing multimodal fusion operators based on the Hadamard product, and show that specific non-trivial instantiations of this generalized fusion operator exhibit superior performance in terms of OpenEnded accuracy on the VQA task. In particular, we introduce Nonlinearity Ensembling, Feature Gating, and post-fusion neural network layers as fusion operator components, culminating in an absolute percentage point improvement of $1.1\%$ on the VQA 2.0 test-dev set over baseline fusion operators, which use the same features as input. We use our findings as evidence that our generalized class of fusion operators could lead to the discovery of even superior task-specific operators when used as a search space in an architecture search over fusion operators.