Search for: All records

Work in computer vision and natural language processing involving images and text has been experiencing explosive growth over the past decade, with a particular boost coming from the neural network revolution. The present volume brings together five research articles from several different corners of the area: multilingual multimodal image description (Frank et al. ), multimodal machine translation (Madhyastha et al. , Frank et al. ), image caption generation (Madhyastha et al. , Tanti et al. ), visual scene understanding (Silberer et al. ), and multimodal learning of high-level attributes (Sorodoc et al. ). In this article, we touch upon all of these topics as we review work involving images and text under the three main headings of image description (Section 2), visually grounded referring expression generation (REG) and comprehension (Section 3), and visual question answering (VQA) (Section 4). 

Peer-reviewed publications and patents serve as important signatures of knowledge generation, and therefore the authors and their organizations can represent agents of intellectual transformation. Accurate tracking of these players enables scholars to follow knowledge evolution. However, while author name disambiguation has been discussed extensively, less is known about the impact of organization name on bibliometric studies. We expand here on the recently defined phenomenon of "onomastic profusion," high-frequency words used in organization names for semantic reasons, and thus contributing a non-random source of error to bibliographic studies. We use the Small Business Innovation Research (SBIR) Phase I awardees of the National Aeronautics and Space Administration (NASA) as a use case in the field of engineering innovation. We find that firms in California or Massachusetts experience a six percent decrease in the likelihood of using the word "Technologies" in their names. Furthermore, use of the words "Research" and "Science" is linked to doubling the number of awards. We illustrate that, in aggregate, firms executing rational strategic naming decisions can create deterministic bibliometric challenges.