Towards Linguistically Generalizable NLP Systems: A Workshop and Shared Task
نویسندگان
چکیده
This paper presents a summary of the first Workshop on Building Linguistically Generalizable Natural Language Processing Systems, and the associated Build It Break It, The Language Edition shared task. The goal of this workshop was to bring together researchers in NLP and linguistics with a shared task aimed at testing the generalizability of NLP systems beyond the distributions of their training data. We describe the motivation, setup, and participation of the shared task, provide discussion of some highlighted results, and discuss lessons learned.
منابع مشابه
Introduction to the Shared Task on Comparing Semantic Representations
Seven groups participated in the STEP 2008 shared task on comparing semantic representations as output by practical wide-coverage NLP systems. Each of this groups developed their own system for producing semantic representations for texts, each in their own semantic formalism. Each group was requested to provide a short sample text, producing a shared task set of seven texts, allowing participa...
متن کاملTÜBİTAK-BİLGEM German-English Machine Translation Systems for W13
This paper describes TÜBİTAK-BİLGEM statistical machine translation (SMT) systems submitted to the Eighth Workshop on Statistical Machine Translation (WMT) shared translation task for German-English language pair in both directions. We implement phrase-based SMT systems with standard parameters. We present the results of using a big tuning data and the effect of averaging tuning weights of diff...
متن کاملTÜBİTAK - BİLGEM German - English Machine Translation Systems for WMT ’ 13
This paper describes TÜBİTAK-BİLGEM statistical machine translation (SMT) systems submitted to the Eighth Workshop on Statistical Machine Translation (WMT) shared translation task for German-English language pair in both directions. We implement phrase-based SMT systems with standard parameters. We present the results of using a big tuning data and the effect of averaging tuning weights of diff...
متن کاملUnsupervised gene function extraction using semantic vectors
UNLABELLED Finding gene functions discussed in the literature is an important task of information extraction (IE) from biomedical documents. Automated computational methodologies can significantly reduce the need for manual curation and improve quality of other related IE systems. We propose an open-IE method for the BioCreative IV GO shared task (subtask b), focused on finding gene function te...
متن کاملTowards Optimal Choice Selection for Improved Hybrid Machine Translation
In recent years, machine translation (MT) research focused on investigating how hybridMT as well as MT combination systems can be designed so that the resulting translations give an improvement over the individual translations. As a first step towards achieving this objective we have developed a parallel corpus with source data and the output of a number of MT systems, annotated with metadata i...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
- CoRR
دوره abs/1711.01505 شماره
صفحات -
تاریخ انتشار 2017