Скорость

This paper describes a "bootstrapping" method which uses a broad-coverage, rule-based parser to compute probabilities while parsing an untagged corpus of NL text, and which then incorporates those probabilities into …...

This paper describes the role of supertagging in a wide-coverage CCG parser which uses a log-linear model to select an analysis. The supertagger reduces the derivation space over which model …...

Trasl´an makes full use of MT during our translation workflow, where the raw output from our Machine Translation (MT) system is passed onto human translators who perform post-editing (if necessary) …...

Post-editing (PE) is still a new activity for many translators. The lack of training, clear and consistent guidelines and inter- national standards may cause difficulties in the transition from translation …...

The usual challenges of transcribing spoken language are compounded for Southern Min (Taiwanese) because it lacks a generally accepted orthography. This study reports the development and testing of software tools …...

We introduce a deep memory network for aspect level sentiment classification. Un- like feature-based SVM and sequential neural models such as LSTM, this approach explic- itly captures the importance of …...

Multiple treebanks annotated under heteroge- neous standards give rise to the research ques- tion of best utilizing multiple resources for im- proving statistical models. Prior research has focused on discrete …...

Given a set of documents from a specific do- main (e.g., medical research journals), how do we automatically build a Knowledge Graph (KG) for that domain? Automatic identifica- tion of …...

AbstractOpen-domain Question Answering models that directly leverage question-answer (QA) pairs, such as closed-book QA (CBQA) models and QA-pair retrievers, show promise in terms of speed and memory compared with conventional …...