End-to-end Knowledge Retrieval with Multi-modal Queries

We investigate knowledge retrieval with multimodal queries, i.e. queries containing information split across image and text inputs, a challenging task that differs from previous work on cross-modal retrieval. We curate a new dataset called ReMuQ 1 for benchmarking progress on this task. ReMuQ requires a system to retrieve knowledge from a large corpus by integrating contents from both text and image queries. We introduce a retriever model "ReViz" that can directly process input text and images to retrieve relevant knowledge in an end-to-end fashion without being dependent on intermediate modules such as object detectors or caption generators. We introduce a new pretraining task that is effective for learning knowledge retrieval with multimodal queries and also improves performance on downstream tasks. We demonstrate superior performance in retrieval on two datasets (ReMuQ and OK-VQA) under zeroshot settings as well as further improvements when finetuned on these datasets. K1: The Empire State Building is a 102-story Art Deco skyscraper in Midtown Manhattan, New York City K2: The 828 metre (2,717 ft) tall Burj Khalifa in Dubai has been the tallest building since 2010. The Burj Khalifa has been classified as megatall. K3: The tallest building in New York is One World Trade Center which rise 1,776 feet (541 m).