VQA-LOL: Visual Question Answering under the Lens of Logic
This addresses robustness in VQA by embedding logical connectives, which is incremental as it builds on existing VQA methods with a focus on logical consistency.
The paper tackled the problem of whether visual question answering (VQA) systems can handle logical compositions of questions, finding that state-of-the-art models struggle with this. The proposed LOL model showed substantial improvement in learning logical compositions while maintaining VQA performance, with specific gains on a constructed benchmark.
Logical connectives and their implications on the meaning of a natural language sentence are a fundamental aspect of understanding. In this paper, we investigate whether visual question answering (VQA) systems trained to answer a question about an image, are able to answer the logical composition of multiple such questions. When put under this \textit{Lens of Logic}, state-of-the-art VQA models have difficulty in correctly answering these logically composed questions. We construct an augmentation of the VQA dataset as a benchmark, with questions containing logical compositions and linguistic transformations (negation, disjunction, conjunction, and antonyms). We propose our {Lens of Logic (LOL)} model which uses question-attention and logic-attention to understand logical connectives in the question, and a novel Fréchet-Compatibility Loss, which ensures that the answers of the component questions and the composed question are consistent with the inferred logical operation. Our model shows substantial improvement in learning logical compositions while retaining performance on VQA. We suggest this work as a move towards robustness by embedding logical connectives in visual understanding.