In production SLU systems, new training data becomes available with time so that ML models need to be updated on a regular basis. Specifically, releasing new features adds new classes of data while the old data remains constant. However, retraining the full model each time from scratch is computationally expensive. To address this problem, we propose to consider production releases from the curriculum learning perspective and to adapt the local-to-global learning (LGL) schedule (Cheng et. al, 2019) for a statistical model that starts with fewer output classes and adds more classes with each iteration. We report experiments for the tasks of intent classification and slot filling in the context of a production voice-assistant. First, we apply the original LGL schedule on our data and then adapt LGL to the production setting where the full data is not available at initial training iterations. We demonstrate that our method improves model error rates by 7.3% and saves up to 25% training time for individual iterations.

True-casing, the task of restoring proper case to (generally) lower case input, is important in downstream tasks and for screen display. In this paper, we investigate truecasing as an in- trinsic task and present several experiments on noisy user queries to a voice-controlled dia- log system. In particular, we compare a rule- based, an n-gram language model (LM) and a recurrent neural network (RNN) approaches, evaluating the results on a German Q&A cor- pus and reporting accuracy for different case categories. We show that while RNNs reach higher accuracy especially on large datasets, character n-gram models with interpolation are still competitive, in particular on mixed- case words where their fall-back mechanisms come into play.

TED-Multilingual Discourse Bank, or TED-MDB, is a multilingual resource where TED-talks are annotated at the discourse level in 6 languages (English, Polish, German, Russian, European Portuguese, and Turkish) following the aims and principles of PDTB. We explain the corpus design criteria, which has three main features: the linguistic characteristics of the languages involved, the interactive nature of TED talks—which led us to annotate Hypophora, and the decision to avoid projection. We report our annotation consistency, and post-annotation alignment experiments, and provide a cross-lingual comparison based on corpus statistics.

Massimo Poesio, Yulia Grishina, Varada Kolhatkar, Nafise Moosavi, Ina Roesiger, Adam Roussel, Fabian Simonjetz, Alexandra Uma, Olga Uryupina, Juntao Yu, Heike Zinsmeister. Proceedings of the First Workshop on Computational Models of Reference, Anaphora and Coreference. 2018.

We report on experiments to validate and extend two language-specific connective databases (German and Italian) using a word-aligned corpus. This is a first step toward constructing a bilingual lexicon on connectives that are connected via their discourse senses.

The CORBON 2017 Shared Task, organised as part of the Coreference Resolution Beyond OntoNotes workshop at EACL 2017, presented a new challenge for multilingual coreference resolution: we offer a projection-based setting in which one is supposed to build a coreference resolver for a new language exploiting little or even no knowledge of it, with our languages of interest being German and Russian. We additionally offer a more traditional setting, targeting the development of a multilingual coreference resolver without any restrictions on the resources and methods used. In this paper, we describe the task setting and provide the results of one participant who successfully completed the task, comparing their results to the closely related previous research. Analysing the task setting and the results, we discuss the major challenges and make suggestions on the future directions of coreference evaluation.

In this paper, we examine the possibility of using annotation projection from multiple sources for automatically obtaining coreference annotations in the target language. We implement a multi-source annotation projection algorithm and apply it on an English-German-Russian parallel corpus in order to transfer coreference chains from two sources to the target side. Operating in two settings – a low-resource and a more linguistically-informed one – we show that automatic coreference transfer could benefit from combining information from multiple languages, and assess the quality of both the extraction and the linking of target coreference mentions.

Although parallel coreference corpora can to a high degree support the development of SMT systems, there are no large-scale parallel datasets available due to the complexity of the annotation task and the variability in annotation schemes. In this study, we exploit an annotation projection method to combine the output of two coreference resolution systems for two different source languages (English, German) in order to create an annotated corpus for a third language (Russian). We show that our technique is superior to projecting annotations from a single source language, and we provide an in-depth analysis of the projected annotations in order to assess the perspectives of our approach.

Anaphoric connectives are event anaphors (or abstract anaphors) that in addition convey a coherence relation holding between the antecedent and the host clause of the connective.Some of them carry an explicitly-anaphoric morpheme, others do not.We analysed the set of German connectives for this property and found that many have an additional nonconnective reading, where they serve as nominal anaphors.Furthermore, many connectives can have multiple senses, so altogether the processing of these words can involve substantial disambiguation.We study the problem for one specific German word, demzufolge, which can be taken as representative for a large group of similar words.

In this paper, we introduce a typology of bridging relations applicable to multiple languages and genres.After discussing our annotation guidelines, we describe annotation experiments on the German part of our parallel coreference corpus and show that our interannotator agreement results are reliable, considering both antecedent selection and relation assignment.In order to validate our theoretical model on other languages, we manually transfer German annotations to the English and Russian sides of the corpus and briefly discuss first results that suggest the promise of our approach.Furthermore, for the complete exploration of extended coreference relations, we exploit an existing near-identity scheme to augment our annotations with near-identity links, and we report on the results.

Building benchmark corpora in the domain of coreference and anaphora resolution is an important task for developing and evaluating NLP systems and models. Our study is aimed at assessing the feasibility of enhancing corpora with information about coreference relations. The annotation procedure includes identification of text segments that are subject to annotation (markables), marking their syntactic heads and identifying coreferential links. Markables are classified according to their morphological, syntactic and reference structure. The annotation is performed manually, providing gold standard data for high-level NLP tasks such as anaphora and coreference resolution. The paper reports on inconsistencies in selecting NPs of various types as markables and their borders, and in ways of constructing anaphoric pairs. We consider the types of NPs missed by annotators, and the discourse and semantic factors that may have affected the annotators’ judgements

Common technologies for automatic coreference resolution require either a language-specific rule set or large collections of manually annotated data, which is typically limited to newswire texts in major languages. This makes it difficult to develop coreference resolvers for a large number of the so-called low-resourced languages. We apply a direct projection algorithm on a multi-genre and multilingual corpus (English, German, Russian) to automatically produce coreference annotations for two target languages without exploiting any linguistic knowledge of the languages. Our evaluation of the projected annotations shows promising results, and the error analysis reveals structural differences of referring expressions and coreference chains for the three languages, which can now be targeted with more linguistically-informed projection algorithms.

A simple conceptual model is employed to investigate events, and break the task of coreference resolution into two steps: semantic class detection and similaritybased matching. With this perspective an algorithm is implemented to cluster event mentions in a large-scale corpus. Results on test data from AQUAINT TimeML, which we annotated manually with coreference links, reveal how semantic conventions vs. information available in the context of event mentions affect decisions in coreference analysis.

ii Introduction Many NLP researchers, especially those not working in the area of discourse processing, tend to equate coreference resolution with the sort of coreference that people did in MUC, ACE, and OntoNotes, having the impression that coreference is a well-worn task owing in part to the large number of papers reporting results on the MUC/ACE/OntoNotes corpora. This is an unfortunate misconception: the previous SemEval 2010 and CoNLL 2012 shared tasks on coreference resolution have largely focused on entity coreference, which constitutes only one of the many kinds of coreference relations that were discussed in theoretical and computational linguistics in the past few decades. In fact, by focusing on entity coreference resolution, NLP researchers have only scratched the surface of the wealth of interesting problems in coreference resolution. The decision to focus on entity coreference resolution was initially made by information extraction (IE) researchers when coreference was selected as one of the tasks in the MUC-6 coreference in 1995. Many interesting kinds of coreference relations, such as bridging and reference to abstract entities, were left out not because they were not important, but because " it was felt that the menu was simply too ambitious ". It turned out that this decision had an important consequence: the progress made in coreference research in the past two decades was largely driven by the availability of coreference-annotated corpora such as MUC, ACE, and OntoNotes, where entity coreference was the focus. Being aware of other fora gathering coreference-related papers (such as LAW, DiscoMT or EVENTS), in 2016 we started a new workshop on the single topic of knowledge-oriented coreference resolution under the name of Coreference Resolution Beyond OntoNotes (CORBON) that would bring together researchers who were interested in under-investigated coreference phenomena, willing to contribute both theoretical and applied computational work on coreference resolution, especially for languages other than English, less-researched forms of coreference and new applications of coreference resolution. The success of the first edition of the workshop (held in conjunction with NAACL HLT 2016 in San Diego, USA) and our intention to verify the role of the Europe-based researchers in the field encouraged us to organise the second edition of the workshop in conjunction with EACL 2017 in Valencia, Spain. Our call attracted 12 submissions (nine from European research institutions and three from India). We are pleased to see that the submissions covered not only a variety of less-studied languages in the …