Unsupervised Induction of Semantic Roles within a Reconstruction-Error Minimization Framework
This provides a language-agnostic approach for semantic role induction, beneficial for NLP tasks in low-resource languages, though it is incremental as it builds on existing reconstruction frameworks.
The paper tackles unsupervised semantic role labeling by jointly training an encoder and a reconstruction model to minimize argument prediction errors, achieving performance comparable to state-of-the-art methods on English and German without using prior linguistic knowledge.
We introduce a new approach to unsupervised estimation of feature-rich semantic role labeling models. Our model consists of two components: (1) an encoding component: a semantic role labeling model which predicts roles given a rich set of syntactic and lexical features; (2) a reconstruction component: a tensor factorization model which relies on roles to predict argument fillers. When the components are estimated jointly to minimize errors in argument reconstruction, the induced roles largely correspond to roles defined in annotated resources. Our method performs on par with most accurate role induction methods on English and German, even though, unlike these previous approaches, we do not incorporate any prior linguistic knowledge about the languages.