RST Discourse Parser for Russian: An Experimental Study of Deep Learning Models

Chistova, Elena; Shelmanov, Artem; Pisarevskaya, Dina; Kobozeva, Maria; Isakov, Vadim; Panchenko, Alexander; Toldova, Svetlana; Smirnov, Ivan V.

doi:10.1007/978-3-030-72610-2_8

Cited by 4 publications

(3 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Named entities are recognized with the SpaCy⁵ ru_core_news_lg model predicting BIO-tags from token embeddings. Discourse structures are produced with the IsaNLP RST⁶ parser for Russian (Chistova et al, 2021). The parser generates trees for each paragraph; we merged these trees with a right-branching multinuclear JOINT relation to construct the full-text RST trees.…”

Section: Instruments For Linguistic Analysismentioning

confidence: 99%

Light Coreference Resolution for Russian with Hierarchical Discourse Features

Chistova,

Smirnov

2023

Computational Linguistics and Intellectual Technologies”

View full text Add to dashboard Cite

Coreference resolution is the task of identifying and grouping mentions referring to the same real-world entity. Previous neural models have mainly focused on learning span representations and pairwise scores for coreference decisions. However, current methods do not explicitly capture the referential choice in the hierarchical discourse, an important factor in coreference resolution. In this study, we propose a new approach that incorporates rhetorical information into neural coreference resolution models. We collect rhetorical featuresfrom automated discourse parses and examine their impact. As a base model, we implement an end-to-end span-based coreference resolver using a partially fine-tuned multilingual entity-aware language model LUKE. We evaluate our method on the RuCoCo-23 Shared Task for coreference resolution in Russian. Our best model employing rhetorical distance between mentions has ranked 1st on the development set (74.6% F1) and 2nd on the test set (73.3% F1) of the Shared Task¹. We hope that our work will inspire further research on incorporating discourse information in neural coreference resolution models.

show abstract

Section: Instruments For Linguistic Analysismentioning

confidence: 99%

Light Coreference Resolution for Russian with Hierarchical Discourse Features

Chistova,

Smirnov

2023

Computational Linguistics and Intellectual Technologies”

View full text Add to dashboard Cite

show abstract

“…In this study, we employ the recent end-to-end RST parsers for English 3 (Zhang et al, 2021) and Russian (Chistova et al, 2020). 3 The models trained on RST-DT corpus.…”

Section: Analyzing Paraphrases From a Discourse Perspectivementioning

confidence: 99%

“…• RST parser for Russian (Chistova et al, 2020), RST-Tace (Wan et al, 2019), rstWeb (Zeldes, 2016), Multilingual DeBERTa v3 (He et al, 2021), spaCy (Honnibal et al, 2020), Evidence Graph framework (Peldszus and Stede, 2015b): MIT License.…”

Section: B2 Opposing Argumentmentioning

confidence: 99%

Discourse-aware text classification for argument mining

Chistova¹,

Smirnov²

2022

Computational Linguistics and Intellectual Technologies

View full text Add to dashboard Cite

We show that using the rhetorical structure automatically generated by the discourse parser is beneficial for paragraph-level argument mining in Russian. First, we improve the structure awareness of the current RST discourse parser for Russian by employing the recent top-down approach for unlabeled tree construction on a paragraph level. Then we demonstrate the utility of this parser in two classification argument mining subtasks of the RuARG-2022 shared task. Our approach leverages a structured LSTM module to compute a text representation that reflects the composition of discourse units in the rhetorical structure. We show that: (i) the inclusion of discourse analysis improves paragraph-level text classification; (ii) a novel TreeLSTM-based approach performs well for the computation of the complex text hidden representation using both a language model and an end-to-end RST parser; (iii) structures predicted by the proposed RST parser reflect the argumentative structures in texts in Russian.

show abstract