LHLibby Hills

Libby Hills

*This is an algorithmically generated author record based on publicly available information.

Libby Hills has a total of 5 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    How AI Works: Reconfiguring Lifelong Learning

    by Ted Palenski, Libby Hills, Sagarika Unnikrishnan, Rebecca Eynon

    Lifelong learning is a current policy focus in many countries, with AI technologies promoted as both the motivation for the need for lifelong learning (due to its assumed role in social change) and as an important way to ‘deliver’ learning across the life course. Such policies tend to be instrumental and technologically deterministic, and there is a need to properly theorize the relationships between AI and lifelong learning to better inform policy and practice. In this paper, we examine the ways that academic communities conceptualize AI and lifelong learning, based on a thematic analysis of existing academic literature in contexts beyond formal education. We identify three groups of research, which vary according to their engagement with theories of learning and AI technology and how AI ‘works’. In group 1 (working AI), AI is assumed to contribute to increased efficiency of humans and learning; in group 2 (working with AI), AI is implemented and conceptualized as a peer or colleague; and in group 3 (reconfiguring AI), AI is viewed as part of a wider reconfiguration of humans and their contexts. This latter group, though least well represented in the literature, holds promise in advancing a postdigital research agenda that focuses not solely on how AI works to increase efficiency, but how people are increasingly working, learning, and living with AI, thus moving beyond exclusively instrumental, economic, and technologically deterministic concerns.


  • Open Access

    Leveraging Human Feedback to Scale Educational Datasets: Combining Crowdworkers and Comparative Judgement

    by Owen Henkel, Libby Hills

    Machine Learning models have many potentially beneficial applications in education settings, but a key barrier to their development is securing enough high-quality, labelled data to train these models. This process has traditionally relied on highly skilled raters using complex, multi-class rubrics, which made labelling expensive and difficult to scale. A more scalable approach would be to use non-expert crowdworkers to evaluate student work, but maintaining high levels of accuracy and inter-rater reliability when using non-expert workers can be challenging. This paper reports on two experiments in which non-expert crowdworkers hired to evaluate (i.e., score) student work and were randomly assigned to one of two conditions: the control, where they were asked to assign a rubrics based score (i.e., a categorical judgement), or the treatment, where they were shown the same student answers, but were asked to decide which of two candidate answers was better (i.e., a comparative/preference-based judgement). We found that using comparative judgement substantially improved inter-rater reliability on both tasks. These results are in-line with well-established literature on the benefits of comparative judgement in the field of educational assessment, as well as with recent trends in artificial intelligence research, where comparative judgement is becoming the preferred method for providing human feedback on model outputs. These results are novel and important in demonstrating the effects of using the combination of comparative judgement and crowdworkers to evaluate educational data