Automatic spell and grammar checking can be done using various system architectures, and large language models have recently been used to solve the task with promising results. Here we describe a new method of creating test data to measure the performance of spell and grammar checkers, including large language models. Three types of test data represent different approaches to evaluation, from basic error detection to error correction with natural language explanations of the corrections made and error severity scores, which is the main novelty of this approach. These additions are especially useful when evaluating large language models. We present a spell and grammar checking test set for Icelandic in which the described approach is applied. The data consists of whole texts instead of discrete sentences, which facilitates evaluating context awareness of models. The resulting test set can be used to compare different spell and grammar checkers and is published under permissive licenses.

Grammatical error correction (GEC) is the task of correcting typos, spelling, punctuation and grammatical issues in text. Approaching the problem as a sequence-to-sequence task, we compare the use of a common subword unit vocabulary and byte-level encoding. Initial synthetic training data is created using an error-generating pipeline, and used for finetuning two subword-level models and one byte-level model. Models are then finetuned further on hand-corrected error corpora, including texts written by children, university students, dyslexic and second-language writers, and evaluated over different error types and error origins. We show that a byte-level model enables higher correction quality than a subword approach, not only for simple spelling errors, but also for more complex semantic, stylistic and grammatical issues. In particular, initial training on synthetic corpora followed by finetuning on a relatively small parallel corpus of real-world errors helps the byte-level model correct a wide range of commonly occurring errors. Our experiments are run for the Icelandic language but should hold for other similar languages, and in particular to morphologically rich ones.

We train several language models for Icelandic, including IceBERT, that achieve state-of-the-art performance in a variety of downstream tasks, including part-of-speech tagging, named entity recognition, grammatical error detection and constituency parsing. To train the models we introduce a new corpus of Icelandic text, the Icelandic Common Crawl Corpus (IC3), a collection of high quality texts found online by targeting the Icelandic top-level-domain .is. Several other public data sources are also collected for a total of 16GB of Icelandic text. To enhance the evaluation of model performance and to raise the bar in baselines for Icelandic, we manually translate and adapt the WinoGrande commonsense reasoning dataset. Through these efforts we demonstrate that a properly cleaned crawled corpus is sufficient to achieve state-of-the-art results in NLP applications for low to medium resource languages, by comparison with models trained on a curated corpus. We further show that initializing models using existing multilingual models can lead to state-of-the-art results for some downstream tasks.

A lack of datasets for spelling and grammatical error correction in Icelandic, along with language-specific issues, has caused a dearth of spell and grammar checking systems for the language. We present the first open-source spell and grammar checking tool for Icelandic, using an error corpus at all stages. This error corpus was in part created to aid in the development of the tool. The system is built with a rule-based tool stack comprising a tokenizer, a morphological tagger, and a parser. For token-level error annotation, tokenization rules, word lists, and a trigram model are used in error detection and correction. For sentence-level error annotation, we use specific error grammar rules in the parser as well as regex-like patterns to search syntax trees. The error corpus gives valuable insight into the errors typically made when Icelandic text is written, and guided each development phase in a test-driven manner. We assess the system’s performance with both automatic and human evaluation, using the test set in the error corpus as a reference in the automatic evaluation. The data in the error corpus development set proved useful in various ways for error detection and correction.

We present Miðeind’s submission for the English→Icelandic and Icelandic→English subsets of the 2021 WMT news translation task. Transformer-base models are trained for translation on parallel data to generate backtranslations teratively. A pretrained mBART-25 model is then adapted for translation using parallel data as well as the last backtranslation iteration. This adapted pretrained model is then used to re-generate backtranslations, and the training of the adapted model is continued.

We present an open-source, wide-coverage context-free grammar (CFG) for Icelandic, and an accompanying parsing system. The grammar has over 5,600 nonterminals, 4,600 terminals and 19,000 productions in fully expanded form, with feature agreement constraints for case, gender, number and person. The parsing system consists of an enhanced Earley-based parser and a mechanism to select best-scoring parse trees from shared packed parse forests. Our parsing system is able to parse about 90% of all sentences in articles published on the main Icelandic news websites. Preliminary evaluation with evalb shows an F-measure of 70.72% on parsed sentences. Our system demonstrates that parsing a morphologically rich language using a wide-coverage CFG can be practical.