Visuo-Lingustic Question Answering (VLQA) Challenge

By Shailaja Keyur Sampat, Yezhou Yang, Chitta Baral. 2020

21 citations recorded by the source corpus.

Understanding images and text together is an important aspect of cognition and building advanced Artificial Intelligence (AI) systems. As a community, we have achieved good benchmarks over language and vision domains separately, however joint reasoning is still a challenge for state-of-the-art computer vision and natural language processing (NLP) systems. We propose a novel task to derive joint inference about a given image-text modality and compile the Visuo-Linguistic Question Answering (VLQA) challenge corpus in a question answering setting. Each dataset item consists of an image and a reading passage, where questions are designed to combine both visual and textual information i.e., ignoring either modality would make the question unanswerable. We first explore the best existing vision-language architectures to solve VLQA subsets and show that they are unable to reason well. We then develop a modular method with slightly better baseline performance, but it is still far behind human performance. We believe that VLQA will be a good benchmark for reasoning over a visuo-linguistic context. The dataset, code and leaderboard is available at https://shailaja183.github.io/vlqa/.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Multimodal Learning, Question Answering & Reading Comprehension

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Chitta Baral
2. Pratyay Banerjee
3. Man Luo
4. Yankai Zeng
5. Swaroop Mishra
6. Bill Byrne
7. Weizhe Lin

In Computer Vision and Pattern Recognition

1. Pratyay Banerjee
2. Man Luo
3. Chitta Baral
4. Yankai Zeng
5. Bill Byrne
6. Weizhe Lin

In Multimodal Learning

1. Pratyay Banerjee
2. Man Luo
3. Chitta Baral
4. Yankai Zeng

In Question Answering & Reading Comprehension

1. Pratyay Banerjee
2. Man Luo
3. Chitta Baral
4. Yankai Zeng
5. Weizhe Lin
6. Bill Byrne

Citations in the corpus

• Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering (2022, neutral): “Different from existing datasets [17, 18, 52], SCIENCEQA has richer domain diversity from three different subjects: natural science, social science, and language science.”
• Weakly-Supervised Visual-Retriever-Reader for Knowledge-based Question Answering (2021, background): “OK-VQA (Marino et al., 2019) covers 11 types of knowledge than previous tasks, such as cooking and food, science and technology, plants and animals, etc. VLQA (Sam-pat et al., 2020) consists of data points of image-passage-question-answer, it is proposed recently to facilitate the research on jointl”