Extra-Propositional Aspects of Meaning in Computational Linguistics Workshop

点击翻译按钮获取中文摘要

This award provides student support for a one-day workshop entitled "Extra-Propositional Aspects of Meaning (ExProM) in Computational Linguistics", to be held collocated with NAACL in Denver on June 5, 2015. Computational semantics has grown significantly in the last decade, but extra-propositional aspects of meaning such as modality, negation and attribution are much less studied: annotations, models and tools are still in their early stages. This workshop brings together students and established researchers interested in defining theoretical frameworks, modelling and implementing real systems to process ExProM, as well analyzing the impact of ExProM in natural language processing applications.Extra-propositional meaning is ubiquitous across domains and genres: it is found in scientific texts, news, social media posts, etc. Humans do not always (want to) say (the whole) truth, and propositional meaning alone disregards significant meaning encoded in everyday texts. The workshop encourages submissions on the following topics (the list is by no means complete): negation: verbal/non-verbal, analytic/synthetic, clausal/subclausal and ordinary/metalinguistic, scope and focus detection; modality: defining and annotating types; factuality: determining factuality changes within and across documents; veridicity and veridicality: measuring author commitment; attribution and perspectives: determining who says what and their purpose; irony and sarcasm; integrating ExProM in the NLP pipeline; and ExProM for NLP applications. The workshop covers these phenomena without targeting any specific domain, and encourages submissions of novel theoretical frameworks and tasks, as well as empirical results on known or novel corpora. Preliminary ideas and results that may lead to active discussion during the event are encouraged. The workshop's format fosters interaction between students and senior researchers, and includes an invited speaker.