BME-TUW at SR’20: Lexical grammar induction for surface realization

no code implementations • MSR (COLING) 2020 • Gábor Recski, Ádám Kovács, Kinga Gémes, Judit Ács, Andras Kornai

We present a system for mapping Universal Dependency structures to raw text which learns to restore word order by training an Interpreted Regular Tree Grammar (IRTG) that establishes a mapping between string and graph operations.

Sentence

Morphosyntactic probing of multilingual BERT models

1 code implementation • 9 Jun 2023 • Judit Acs, Endre Hamerlik, Roy Schwartz, Noah A. Smith, Andras Kornai

We introduce an extensive dataset for multilingual probing of morphological information in language models (247 tasks across 42 languages from 10 families), each consisting of a sentence with a target word and a morphological tag as the desired label, derived from the Universal Dependencies treebanks.

Sentence TAG

Subword Pooling Makes a Difference

1 code implementation • EACL 2021 • Judit {\'A}cs, {\'A}kos K{\'a}d{\'a}r, Andras Kornai

For POS tagging both of these strategies perform poorly and the best choice is to use a small LSTM over the subwords.

NER POS +1

The Role of Interpretable Patterns in Deep Learning for Morphology

1 code implementation • 8 Dec 2020 • Judit Acs, Andras Kornai

By training the models on the same source but different target, we can compare what subwords are important for different tasks and how they relate to each other.

Lemmatization Morphological Analysis

BMEAUT at SemEval-2020 Task 2: Lexical Entailment with Semantic Graphs

no code implementations • SEMEVAL 2020 • {\'A}d{\'a}m Kov{\'a}cs, Kinga G{\'e}mes, Andras Kornai, G{\'a}bor Recski

In this paper we present a novel rule-based, language independent method for determining lexical entailment relations using semantic representations built from Wiktionary definitions.

Lexical Entailment Semantic Parsing +1

Better Together: Modern Methods Plus Traditional Thinking in NP Alignment

no code implementations • LREC 2020 • {\'A}d{\'a}m Kov{\'a}cs, Judit {\'A}cs, Andras Kornai, G{\'a}bor Recski

We study a typical intermediary task to Machine Translation, the alignment of NPs in the bitext.

Lemmatization Machine Translation +1

BME-UW at SRST-2019: Surface realization with Interpreted Regular Tree Grammars

no code implementations • WS 2019 • {\'A}d{\'a}m Kov{\'a}cs, Evelin {\'A}cs, Judit {\'A}cs, Andras Kornai, G{\'a}bor Recski

The Surface Realization Shared Task involves mapping Universal Dependency graphs to raw text, i. e. restoring word order and inflection from a graph of typed, directed dependencies between lemmas.

Semantic Parsing

Naive probability

no code implementations • 20 May 2019 • Zalan Gyenis, Andras Kornai

We describe a rational, but low resolution model of probability.