Learning Disentangled Representations of Texts with Application to Biomedical Abstracts
This addresses the need for interpretable and transferable text embeddings in biomedical and review domains, though it is incremental as it builds on existing disentanglement methods.
The paper tackles the problem of learning disentangled text representations to encode distinct aspects like populations, interventions, and outcomes in biomedical abstracts, resulting in effective aspect-specific retrieval and generalization to other multi-aspect corpora.
We propose a method for learning disentangled representations of texts that code for distinct and complementary aspects, with the aim of affording efficient model transfer and interpretability. To induce disentangled embeddings, we propose an adversarial objective based on the (dis)similarity between triplets of documents with respect to specific aspects. Our motivating application is embedding biomedical abstracts describing clinical trials in a manner that disentangles the populations, interventions, and outcomes in a given trial. We show that our method learns representations that encode these clinically salient aspects, and that these can be effectively used to perform aspect-specific retrieval. We demonstrate that the approach generalizes beyond our motivating application in experiments on two multi-aspect review corpora.