The paper deals with the pilot version of the first RST discourse treebank for Russian. The project started in 2016. At present, the treebank consists of sixty news texts annotated for rhetorical relations according to RST scheme. However, this scheme was slightly modified in order to achieve higher inter-annotator agreement score. During the annotation procedure, we also registered the discourse connectives of different types and mapped them onto the corresponding rhetoric relations. In present paper, we discuss our experience of RST scheme adaptation for Russian news texts. Besides, we report on the distribution of the most frequent discourse connectives in our corpus.

A broad goal in natural language processing (NLP) is to develop a system that has the capacity to process any natural language. Most systems, however, are developed using data from just one language such as English. The SIG-MORPHON 2020 shared task on morphological reinflection aims to investigate systems' ability to generalize across typologically distinct languages, many of which are low resource. Systems were developed using data from 45 languages and just 5 language families, fine-tuned with data from an additional 45 languages and 10 language families (13 in total), and evaluated on all 90 languages. A total of 22 systems (19 neural) from 10 teams were submitted to the task. All four winning systems were neural (two monolingual transformers and two massively multilingual RNNbased models with gated attention). Most teams demonstrate utility of data hallucination and augmentation, ensembles, and multilingual training for low-resource languages. Nonneural learners and manually designed grammars showed competitive and even superior performance on some languages (such as Ingrian, Tajik, Tagalog, Zarma, Lingala), especially with very limited data. Some language families (Afro-Asiatic, Niger-Congo, Turkic) were relatively easy for most systems and achieved over 90% mean accuracy while others were more challenging.

This paper presents a quantitative and qualitative error analysis of Russian anaphora resolvers which participated in the RU-EVAL event. Its aim is to identify and characterize a set of challenging errors common to stateof-the-art systems dealing with Russian. We examined three types of pronouns: 3rd person pronouns, reflexive and relative pronouns. The investigation has shown that a high level of grammatical ambiguity, specific features of reflexive pronouns, free word order and special cases of non-referential pronouns in Russian impact the quality of anaphora resolution systems. Error analysis reveals some specific features of anaphora resolution for morphologically rich and free word order languages with a lack of gold standard resources.

No abstract

No abstract