This study examines the lexical and syntactic interventions of human and LLM proofreading aimed at improving overall intelligibility in identical second language writings, and evaluates the consistency of outcomes across three LLMs (ChatGPT-4o, Llama3.1-8b, Deepseek-r1-8b). Findings show that both human and LLM proofreading enhance bigram lexical features, which may contribute to better coherence and contextual connectedness between adjacent words. However, LLM proofreading exhibits a more generative approach, extensively reworking vocabulary and sentence structures, such as employing more diverse and sophisticated vocabulary and incorporating a greater number of adjective modifiers in noun phrases. The proofreading outcomes are highly consistent in major lexical and syntactic features across the three models.

The present study extends recent work on Universal Dependencies annotations for second-language (L2) Korean by introducing a semi-automated framework that identifies morphosyntactic constructions from XPOS sequences and aligns those constructions with corresponding UPOS categories. We also broaden the existing L2-Korean corpus by annotating 2,998 new sentences from argumentative essays. To evaluate the impact of XPOS-UPOS alignments, we fine-tune L2-Korean morphosyntactic analysis models on datasets both with and without these alignments, using two NLP toolkits. Our results indicate that the aligned dataset not only improves consistency across annotation layers but also enhances morphosyntactic tagging and dependency-parsing accuracy, particularly in cases of limited annotated data.

We expand the second language (L2) Korean Universal Dependencies (UD) treebank with 5,454 manually annotated sentences. The annotation guidelines are also revised to better align with the UD framework. Using this enhanced treebank, we fine-tune three Korean language models—Stanza, spaCy, and Trankit—and evaluate their performance on in-domain and out-of-domain L2-Korean datasets. The results show that fine-tuning significantly improves their performance across various metrics, thus highlighting the importance of using well-tailored L2 datasets for fine-tuning first-language-based, general-purpose language models for the morphosyntactic analysis of L2 data.

This study evaluates the effectiveness of pre-trained language models in identifying argument structure constructions, important for modeling both first and second language learning. We examine three methodologies: (1) supervised training with RoBERTa using a gold-standard ASC treebank, including by-tag accuracy evaluation for sentences from both native and non-native English speakers, (2) prompt-guided annotation with GPT-4, and (3) generating training data through prompts with GPT-4, followed by RoBERTa training. Our findings indicate that RoBERTa trained on gold-standard data shows the best performance. While data generated through GPT-4 enhances training, it does not exceed the benchmarks set by gold-standard data.

We introduce a detailed annotation scheme for argument structure constructions (ASCs) along with a manually annotated ASC treebank. This treebank encompasses 10,204 sentences from both first (5,936) and second language English datasets (1,948 for written; 2,320 for spoken). We detail the annotation process and evaluate inter-annotation agreement for overall and each ASC category.

We introduce a manually annotated syntactic treebank based on Universal Dependencies, derived from the written data of second language (L2) Korean learners. In developing this new dataset, we critically evaluated previous works and revised the annotation guidelines to better reflect the linguistic properties of Korean and the characteristics of L2 learners. The L2 Korean treebank encompasses 7,530 sentences (66,982 words; 129,333 morphemes) and is publicly available at: https://github.com/NLPxL2Korean/L2KW-corpus.

We introduce the Korean-Learner-Morpheme (KLM) corpus, a manually annotated dataset consisting of 129,784 morphemes from second language (L2) learners of Korean, featuring morpheme tokenization and part-of-speech (POS) tagging. We evaluate the performance of four Korean morphological analyzers in tokenization and POS tagging on the L2- Korean corpus. Results highlight the analyzers’ reduced performance on L2 data, indicating the limitation of advanced deep-learning models when dealing with L2-Korean corpora. We further show that fine-tuning one of the models with the KLM corpus improves its accuracy of tokenization and POS tagging on L2-Korean dataset.

In this paper we introduce a freely available treebank that includes argument structure construction (ASC) annotation. We then use the treebank to train probabilistic annotation models that rely on verb lemmas and/ or syntactic frames. We also use the treebank data to train a highly accurate transformer-based annotation model (F1 = 91.8%). Future directions for the development of the treebank and annotation models are discussed.

This study investigates the extent to which currently available morpheme parsers/taggers apply to lesser-studied languages and language-usage contexts, with a focus on second language (L2) Korean. We pursue this inquiry by (1) training a neural-network model (pre-trained on first language [L1] Korean data) on varying L2 datasets and (2) measuring its morpheme parsing/POS tagging performance on L2 test sets from both the same and different sources of the L2 train sets. Results show that the L2 trained models generally excel in domain-specific tokenization and POS tagging compared to the L1 pre-trained baseline model. Interestingly, increasing the size of the L2 training data does not lead to improving model performance consistently.