Discourse structure is integral to understanding a text and is helpful in many NLP tasks. Learning latent representations of discourse is an attractive alternative to acquiring expensive labeled discourse data. Liu and Lapata (2018) propose a structured attention mechanism for text classification that derives a tree over a text, akin to an RST discourse tree. We examine this model in detail, and evaluate on additional discourse-relevant tasks and datasets, in order to assess whether the structured attention improves performance on the end task and whether it captures a text's discourse structure. We find the learned latent trees have little to no structure and instead focus on lexical cues; even after obtaining more structured trees with proposed model modifications, the trees are still far from capturing discourse structure when compared to discourse dependency trees from an existing discourse parser. Finally, ablation studies show the structured attention provides little benefit, sometimes even hurting performance. 1

Parkinson’s Disease can be understood as a disorder of motor habits. A prediction of this theory is that early stage Parkinson’s patients will display fewer errors caused by interference from previously over-learned behaviours. We test this prediction in the domain of skilled typing, where actions are easy to record and errors easy to identify. We describe a method for categorizing errors as simple motor errors or habit-driven errors. We test Spanish and English participants with and without Parkinson’s, and show that indeed patients make fewer habit errors than healthy controls, and, further, that classification of error type increases the accuracy of discriminating between patients and healthy controls. As well as being a validation of a theory-led prediction, these results offer promise for automated, enhanced and early diagnosis of Parkinson’s Disease.

Parkinson’s Disease can be understand as a disorder of motor habits. A prediction of this theory is that early stage Parkinson’s patients will display fewer errors caused by interference from previously over-learned behaviours. We test this prediction in the domain of skilled typing, where actions are easy to record and errors easy to identify. We describe a method for categorising errors as simple motor errors or habit-driven errors. We test Spanish and English participants with and without Parkinson’s, and show that indeed patients make fewer habit errors than healthy controls, and, further, that classification of error type increases the accuracy of discriminating between patients and healthy controls. As well as being a validation of a theory-led prediction, these results offer promise for automated, enhanced and early diagnosis of Parkinson’s Disease.

Discourse signals are often implicit, leaving it up to the interpreter to draw the required inferences. At the same time, discourse is embedded in a social context, meaning that interpreters apply their own assumptions and beliefs when resolving these inferences, leading to multiple, valid interpretations. However, current discourse data and frameworks ignore the social aspect, expecting only a single ground truth. We present the first discourse dataset with multiple and subjective interpretations of English conversation in the form of perceived conversation acts and intents. We carefully analyze our dataset and create computational models to (1) confirm our hypothesis that taking into account the bias of the interpreters leads to better predictions of the interpretations, (2) and show disagreements are nuanced and require a deeper understanding of the different contextual factors. We share our dataset and code at http://github.com/ elisaF/subjective_discourse.

Performing systematic reviews is a critical yet manual, labor-intensive step in evidencebased medicine. Automating systematic reviews is an active area of research, requiring innovations in machine learning and computational linguistics. We examine how coreference resolution can aid in identifying the arms of a study, an often overlooked piece of information needed to synthesize the results in a systematic review. A classification model 1 that performs better with the coreference features supports the intuition that coreference is able to capture the discourse salience of arms. We note that control arms do not benefit as much from these features.