Image captioning is an important task for benchmarking visual reasoning and for enabling accessibility for people with vision impairments. However, as in many machine learning settings, social biases can influence image captioning in undesirable ways. In this work, we study bias propagation pathways within image captioning, focusing specifically on the COCO dataset. Prior work has analyzed gender bias in captions using automatically-derived gender labels; here we examine racial and intersectional biases using manual annotations. Our first contribution is in annotating the perceived gender and skin color of 28,315 of the depicted people after obtaining IRB approval. Using these annotations, we compare racial biases present in both manual and automatically-generated image captions. We demonstrate differences in caption performance, sentiment, and word choice between images of lighter versus darker-skinned people. Further, we find the magnitude of these differences to be greater in modern captioning systems compared to older ones, thus leading to concerns that without proper consideration and mitigation these differences will only become increasingly prevalent. Code and data is available at https://princetonvisualai. github.io/imagecaptioning-bias/.
As teenage use of social media platform continues to proliferate, so do concerns about teenage privacy and safety online. Prior work has established that privacy on networked publics, such as social media, is complex, requiring users to navigate not only the technical affordances on the platform but also interpersonal relationships and social norms. We investigate how teenagers think about privacy on the popular imagesharing platform, Instagram. We draw on an online survey (N=144) and semi-structured interviews (N=21) with teenagers, ages 13-19, to gain a better understanding how teenagers configure privacy on the popular image-sharing platform Instagram and why they make these privacy decisions. Finally, based on our findings, we provide design recommendations towards the design of better privacy controls for promoting teenage safety online.CCS Concepts: • Human-centered computing → Empirical studies in collaborative and social computing; Empirical studies in HCI; • Security and privacy → Social aspects of security and privacy.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.