In this project, we have investigated the use of advanced machine learning methods, specifically fine-tuned large language models, for pre-annotating data for a lexical extension task, namely adding descriptive words (verbs) to an existing (but incomplete, as of yet) ontology of event types. Several research questions have been focused on, from the investigation of a possible heuristics to provide at least hints to annotators which verbs to include and which are outside the current version of the ontology, to the possible use of the automatic scores to help the annotators to be more efficient in finding a threshold for identifying verbs that cannot be assigned to any existing class and therefore they are to be used as seeds for a new class. We have also carefully examined the correlation of the automatic scores with the human annotation. While the correlation turned out to be strong, its influence on the annotation proper is modest due to its near linearity, even though the mere fact of such pre-annotation leads to relatively short annotation times.
We describe the winning submission to the CRAC 2022 Shared Task on Multilingual Coreference Resolution. Our system first solves mention detection and then coreference linking on the retrieved spans with an antecedent-maximization approach, and both tasks are fine-tuned jointly with shared Transformer weights. We report results of finetuning a wide range of pretrained models. The center of this contribution are fine-tuned multilingual models. We found one large multilingual model with sufficiently large encoder to increase performance on all datasets across the board, with the benefit not limited only to the underrepresented languages or groups of typologically relative languages. The source code is available at https://github.com/ufal/crac2022-corpipe.
We introduce a large and diverse Czech corpus annotated for grammatical error correction (GEC) with the aim to contribute to the still scarce data resources in this domain for languages other than English. The Grammar Error Correction Corpus for Czech (GECCC) offers a variety of four domains, covering error distributions ranging from high error density essays written by non-native speakers, to website texts, where errors are expected to be much less common. We compare several Czech GEC systems, including several Transformer-based ones, setting a strong baseline to future research. Finally, we meta-evaluate common GEC metrics against human judgments on our data. We make the new Czech GEC corpus publicly available under the CC BY-SA 4.0 license at http://hdl.handle.net/11234/1-4639.
Sensitivity of deep-neural models to input noise is known to be a challenging problem. In NLP, model performance often deteriorates with naturally occurring noise, such as spelling errors. To mitigate this issue, models may leverage artificially noised data. However, the amount and type of generated noise has so far been determined arbitrarily. We therefore propose to model the errors statistically from grammatical-error-correction corpora. We present a thorough evaluation of several state-of-the-art NLP systems’ robustness in multiple languages, with tasks including morpho-syntactic analysis, named entity recognition, neural machine translation, a subset of the GLUE benchmark and reading comprehension. We also compare two approaches to address the performance drop: a) training the NLP models with noised data generated by our framework; and b) reducing the input noise with external system for natural language correction. The code is released at https://github.com/ufal/kazitext.
We propose a character-based non-autoregressive GEC approach, with automatically generated character transformations. Recently, per-word classification of correction edits has proven an efficient, parallelizable alternative to current encoder-decoder GEC systems. We show that word replacement edits may be suboptimal and lead to explosion of rules for spelling, diacritization and errors in morphologically rich languages, and propose a method for generating character transformations from GEC corpus. Finally, we train character transformation models for Czech, German and Russian, reaching solid results and dramatic speedup compared to autoregressive systems. The source code is released at https://github.com/ufal/wnut2021_character_transformations_gec.
We present our contribution to the EvaLatin shared task, which is the first evaluation campaign devoted to the evaluation of NLP tools for Latin. We submitted a system based on UDPipe 2.0, one of the winners of the CoNLL 2018 Shared Task, The 2018 Shared Task on Extrinsic Parser Evaluation and SIGMORPHON 2019 Shared Task. Our system places first by a wide margin both in lemmatization and POS tagging in the open modality, where additional supervised data is allowed, in which case we utilize all Universal Dependency Latin treebanks. In the closed modality, where only the EvaLatin training data is allowed, our system achieves the best performance in lemmatization and in classical subtask of POS tagging, while reaching second place in cross-genre and cross-time settings. In the ablation experiments, we also evaluate the influence of BERT and XLM-RoBERTa contextualized embeddings, and the treebank encodings of the different flavors of Latin treebanks.
We propose two neural network architectures for nested named entity recognition (NER), a setting in which named entities may overlap and also be labeled with more than one label. We encode the nested labels using a linearized scheme. In our first proposed approach, the nested labels are modeled as multilabels corresponding to the Cartesian product of the nested labels in a standard LSTM-CRF architecture. In the second one, the nested NER is viewed as a sequence-to-sequence problem, in which the input sequence consists of the tokens and output sequence of the labels, using hard attention on the word whose label is being predicted. The proposed methods outperform the nested NER state of the art on four corpora: ACE-2004, ACE-2005, GENIA and Czech CNEC. We also enrich our architectures with the recently published contextual embeddings: ELMo, BERT and Flair, reaching further improvements for the four nested entity corpora. In addition, we report flat NER state-of-the-art results for CoNLL-2002 Dutch and Spanish and for CoNLL-2003 English.
We present a system description of our contribution to the CoNLL 2019 shared task, CrossFramework Meaning Representation Parsing (MRP 2019). The proposed architecture is our first attempt towards a semantic parsing extension of the UDPipe 2.0, a lemmatization, POS tagging and dependency parsing pipeline. For the MRP 2019, which features five formally and linguistically different approaches to meaning representation (DM, PSD, EDS, UCCA and AMR), we propose a uniform, language and framework agnostic graph-tograph neural network architecture. Without any knowledge about the graph structure, and specifically without any linguistically or framework motivated features, our system implicitly models the meaning representation graphs. After fixing a human error (we used earlier incorrect version of provided test set analyses), our submission would score third in the competition evaluation. The source code of our system is available at https://github.com/ufal/mrpipe-conll2019.
We present our contribution to the SIGMORPHON 2019 Shared Task: Crosslinguality and Context in Morphology, Task 2: contextual morphological analysis and lemmatization. We submitted a modification of the UDPipe 2.0, one of best-performing systems of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies and an overall winner of the The 2018 Shared Task on Extrinsic Parser Evaluation. As our first improvement, we use the pretrained contextualized embeddings (BERT) as additional inputs to the network; secondly, we use individual morphological features as regularization; and finally, we merge the selected corpora of the same language. In the lemmatization task, our system exceeds all the submitted systems by a wide margin with lemmatization accuracy 95.78 (second best was 95.00, third 94.46). In the morphological analysis, our system placed tightly second: our morphological analysis accuracy was 93.19, the winning system’s 93.23.
Many natural language processing tasks, including the most advanced ones, routinely start by several basic processing steps – tokenization and segmentation, most likely also POS tagging and lemmatization, and commonly parsing as well. A multilingual pipeline performing these steps can be trained using the Universal Dependencies project, which contains annotations of the described tasks for 50 languages in the latest release UD 2.0. We present an update to UDPipe, a simple-to-use pipeline processing CoNLL-U version 2.0 files, which performs these tasks for multiple languages without requiring additional external data. We provide models for all 50 languages of UD 2.0, and furthermore, the pipeline can be trained easily using data in CoNLL-U format. UDPipe is a standalone application in C++, with bindings available for Python, Java, C# and Perl. In the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, UDPipe was the eight best system, while achieving low running times and moderately sized models.
Automatic natural language processing of large texts often presents recurring challenges in multiple languages: even for most advanced tasks, the texts are first processed by basic processing steps – from tokenization to parsing. We present an extremely simple-to-use tool consisting of one binary and one model (per language), which performs these tasks for multiple languages without the need for any other external data. UDPipe, a pipeline processing CoNLL-U-formatted files, performs tokenization, morphological analysis, part-of-speech tagging, lemmatization and dependency parsing for nearly all treebanks of Universal Dependencies 1.2 (namely, the whole pipeline is currently available for 32 out of 37 treebanks). In addition, the pipeline is easily trainable with training data in CoNLL-U format (and in some cases also with additional raw corpora) and requires minimal linguistic knowledge on the users’ part. The training code is also released.
In recent years, considerable attention has been dedicated to language modeling methods in information retrieval. Although these approaches generally allow exploitation of any type of language model, most of the published experiments were conducted with a classical n-gram model, usually limited only to unigrams. A few works exploiting syntax in information retrieval can be cited in this context, but significant contribution of syntax based language modeling for information retrieval is yet to be proved. In this paper, we propose, implement, and evaluate an enrichment of language model employing syntactic dependency information acquired automatically from both documents and queries. Our experiments are conducted on Czech which is a morphologically rich language and has a considerably free word order, therefore a syntactic language model is expected to contribute positively to the unigram and bigram language model based on surface word order. By testing our model on the Czech test collection from Cross Language Evaluation Forum 2007 Ad-Hoc track, we show positive contribution of using dependency syntax in this context.