We study the problem of zero-shot exercise retrieval in the context of online language learning, to give learners the ability to explicitly request personalized exercises via natural language. Using real-world data collected from language learners, we observe that vector similarity approaches poorly capture the relationship between exercise content and the language that learners use to express what they want to learn. This semantic gap between queries and content dramatically reduces the effectiveness of general-purpose retrieval models pretrained on large scale information retrieval datasets like MS MARCO [2]. We leverage the generative capabilities of large language models to bridge the gap by synthesizing hypothetical exercises based on the learner’s input, which are then used to search for relevant exercises. Our approach, which we call mHyER, overcomes three challenges: (1) lack of relevance labels for training, (2) unrestricted learner input content, and (3) low semantic similarity between input and retrieval candidates. mHyER outperforms several strong baselines on two novel benchmarks created from crowdsourced data and publicly available data.

Learning engineering adds tools and processes to learning platforms to support improvement research. One kind of tool is A/B testing, which is common in large software companies and also represented academically at conferences like the Annual Conference on Digital Experimentation (CODE). A number of A/B testing systems focused on educational applications have arisen recently, including UpGrade and E-TRIALS. A/B testing can be part of the puzzle of how to improve educational platforms, and yet challenging issues in education go beyond the generic paradigm. For example, the importance of teachers and instructors to learning means that students are not only connecting with software as individuals, but also as part of a shared classroom experience. Further, learning in topics like mathematics can be highly dependent on prior learning, and thus A or B may not be better overall, but only in interaction with prior knowledge. In response, a set of learning platforms is opening their systems to improvement research by instructors and/or third-party researchers, with specific supports necessary for education-specific research designs. This workshop will explore how A/B testing in educational contexts is different, how learning platforms are opening up new possibilities, and how these empirical approaches can be used to drive powerful gains in student learning. It will also discuss forthcoming opportunities for funding to conduct platform-enabled learning research.

The emerging discipline of Learning Engineering is focused on putting into place tools and processes that use the science of learning as a basis for improving educational outcomes. An important part of Learning Engineering focuses on improving the effectiveness of educational software. In many software domains, A/B testing has become a prominent technique to achieve the software's goals. Many large companies (Amazon, Google, Facebook, etc.) run thousands of AB tests and present at the Annual Conference on Digital Experimentation (CODE), but that venue is too broad to address AB testing issues specific to EdTech platforms. We see a need to address issues with running large-scale A/B tests within the educational context, where the use of A/B testing lags other industries. This workshop will explore ways in which A/B testing in educational contexts differs from other domains and proposals to overcome current challenges so that this approach can become a more useful tool in the learning engineer's toolbox.

To ascertain the importance of phonetic information in the form of phonological distinctive features for the purpose of segmentlevel phonotactic acquisition, we compare the performance of two recurrent neural network models of phonotactic learning: one that has access to distinctive features at the start of the learning process, and one that does not.Though the predictions of both models are significantly correlated with human judgments of non-words, the feature-naive model significantly outperforms the feature-aware one in terms of probability assigned to a held-out test set of English words, suggesting that distinctive features are not obligatory for learning phonotactic patterns at the segment level.

We present a method to jointly learn fea-tures and weights directly from distri-butional data in a log-linear framework. Specifically, we propose a non-parametric Bayesian model for learning phonologi-cal markedness constraints directly from the distribution of input-output mappings in an Optimality Theory (OT) setting. The model uses an Indian Buffet Process prior to learn the feature values used in the log-linear method, and is the first algorithm for learning phonological constraints with-out presupposing constraint structure. The model learns a system of constraints that explains observed data as well as the phonologically-grounded constraints of a standard analysis, with a violation struc-ture corresponding to the standard con-straints. These results suggest an alterna-tive data-driven source for constraints in-stead of a fully innate constraint set. 1