This study evaluates Large Language Models’ (LLMs) ability to simulate non-native-like English use observed in human second language (L2) learners interfered with by their native first language (L1). In dialogue-based interviews, we prompt LLMs to mimic L2 English learners with specific L1s (e.g., Japanese, Thai, Urdu) across seven languages, comparing their outputs to real L2 learner data. Our analysis examines L1-driven linguistic biases, such as reference word usage and avoidance behaviors, using information-theoretic and distributional density measures. Results show that modern LLMs (e.g., Qwen2.5, LLAMA3, DeepseekV3, GPT 4o) replicate L1-dependent patterns observed in human L2 data, with distinct influences from various languages (e.g., Japanese, Korean, and Mandarin significantly affect tense agreement, and Urdu influences noun-verb collocations). Our results reveal LLMs’ potential for L2 dialogue generation and evaluation for future educational applications.

The performance and usability of Large-Language Models (LLMs) are driving their use in explanation generation tasks. However, despite their widespread adoption, LLM explanations have been found to be unreliable, making it difficult for users to distinguish good from bad explanations. To address this issue, we present Rubrik’s CUBE–an education-inspired rubric and a dataset of 26k explanations, written and later quality-annotated using the rubric by both humans and six open- and closed-source LLMs. The CUBE dataset focuses on two reasoning and two language tasks, providing the necessary diversity for us to effectively test our proposed rubric. Using Rubrik, we find that explanations are influenced by both task and perceived difficulty. Low quality stems primarily from a lack of conciseness in LLM-generated explanations, rather than cohesion and word choice. The full dataset, rubric, and code are available at https://github.com/RubriksCube/rubriks_cube.

Knowing which words language learners struggle with is crucial for developing personalised education technologies. In this paper, we advocate for the novel task of “dictionary look-up prediction” as a means for evaluating the complexity of words in reading tasks. We release the Dictionary Look-Up development dataset (DLU-dev) and the Dialogue Dictionary Look-Up dataset (D-DLU), which is based on chatbot dialogues. We demonstrate that dictionary look-up is a challenging task for LLMs (results are presented for LLaMA, Gemma, and Longformer models). We explore finetuning with the ROC* loss function as a more appropriate loss for this task than the commonly used Binary Cross Entropy (BCE). We show that a feature-based model outperforms the LLMs. Finally, we investigate the transfer between DLU and the related tasks of Complex Word Identification (CWI) and Semantic Error Prediction (SEP), establishing new state-of-the-art results for SEP.

With the rapid development of large language models (LLMs), LLM-as-a-judge has emerged as a widely adopted approach for text quality evaluation, including hallucination evaluation. While previous studies have focused exclusively on single-context evaluation (e.g., discourse faithfulness or world factuality), real-world hallucinations typically involve mixed contexts, which remains inadequately evaluated. In this study, we use summarization as a representative task to comprehensively evaluate LLMs’ capability in detecting mixed-context hallucinations, specifically distinguishing between factual and non-factual hallucinations. Through extensive experiments across direct generation and retrieval-based models of varying scales, our main observations are: (1) LLMs’ intrinsic knowledge introduces inherent biases in hallucination evaluation; (2) These biases particularly impact the detection of factual hallucinations, yielding a significant performance bottleneck; and (3) the fundamental challenge lies in effective knowledge utilization, balancing between LLMs’ intrinsic knowledge and external context for accurate mixed-context hallucination evaluation.

With the rise of globalisation, code-switching (CSW) has become a ubiquitous part of multilingual conversation, posing new challenges for natural language processing (NLP), especially in Grammatical Error Correction (GEC). This work explores the complexities of applying GEC systems to CSW texts. Our objectives include evaluating the performance of state-of-the-art GEC systems on an authentic CSW dataset from English as a Second Language (ESL) learners, exploring synthetic data generation as a solution to data scarcity, and developing a model capable of correcting grammatical errors in monolingual and CSW texts. We generated synthetic CSW GEC data, resulting in one of the first substantial datasets for this task, and showed that a model trained on this data is capable of significant improvements over existing systems. This work targets ESL learners, aiming to provide educational technologies that aid in the development of their English grammatical correctness without constraining their natural multilingualism.

Thanks to recent advances in generative AI, we are able to prompt large language models (LLMs) to produce texts which are fluent and grammatical. In addition, it has been shown that we can elicit attempts at grammatical error correction (GEC) from LLMs when prompted with ungrammatical input sentences. We evaluate how well LLMs can perform at GEC by measuring their performance on established benchmark datasets. We go beyond previous studies, which only examined GPT* models on a selection of English GEC datasets, by evaluating seven open-source and three commercial LLMs on four established GEC benchmarks. We investigate model performance and report results against individual error types. Our results indicate that LLMs do not always outperform supervised English GEC models except in specific contexts – namely commercial LLMs on benchmarks annotated with fluency corrections as opposed to minimal edits. We find that several open-source models outperform commercial ones on minimal edit benchmarks, and that in some settings zero-shot prompting is just as competitive as few-shot prompting.

Lexical Simplification (LS) aims to simplify text at the lexical level. Existing methods rely heavily on annotated data, making it challenging to apply in low-resource scenarios. In this paper, we propose a novel LS method without parallel corpora. This method employs an Adversarial Editing System with guidance from a confusion loss and an invariance loss to predict lexical edits in the original sentences. Meanwhile, we introduce an innovative LLM-enhanced loss to enable the distillation of knowledge from Large Language Models (LLMs) into a small-size LS system. From that, complex words within sentences are masked and a Difficulty-aware Filling module is crafted to replace masked positions with simpler words. At last, extensive experimental results and analyses on three benchmark LS datasets demonstrate the effectiveness of our proposed method.

In this study, we evaluated the performance of the state-of-the-art sequence tagging grammar error detection and correction model (SeqTagger) using Japanese university students’ writing samples. With an automatic annotation toolkit, ERRANT, we first evaluated SeqTagger’s performance on error correction with human expert correction as the benchmark. Then a human-annotated approach was adopted to evaluate Seqtagger’s performance in error detection using a subset of the writing dataset. Results indicated a precision of 63.66% and a recall of 20.19% for error correction in the full dataset. For the subset, after manual exclusion of irrelevant errors such as semantic and mechanical ones, the model shows an adjusted precision of 97.98% and an adjusted recall of 42.98% for error detection, indicating the model’s high accuracy but also its conservativeness. Thematic analysis on errors undetected by the model revealed that determiners and articles, especially the latter, were predominant. Specifically, in terms of context-independent errors, the model occasionally overlooked basic ones and faced challenges with overly erroneous or complex structures. Meanwhile, context-dependent errors, notably those related to tense and noun number, as well as those possibly influenced by the students’ first language (L1), remained particularly challenging.

Code-switching (CSW) is a common phenomenon among multilingual speakers where multiple languages are used in a single discourse or utterance. Mixed language utterances may still contain grammatical errors however, yet most existing Grammar Error Correction (GEC) systems have been trained on monolingual data and not developed with CSW in mind. In this work, we conduct the first exploration into the use of GEC systems on CSW text. Through this exploration, we propose a novel method of generating synthetic CSW GEC datasets by translating different spans of text within existing GEC corpora. We then investigate different methods of selecting these spans based on CSW ratio, switch-point factor and linguistic constraints, and identify how they affect the performance of GEC systems on CSW text. Our best model achieves an average increase of 1.57 F0.5 across 3 CSW test sets (English-Chinese, English-Korean and English-Japanese) without affecting the model’s performance on a monolingual dataset. We furthermore discovered that models trained on one CSW language generalise relatively well to other typologically similar CSW languages.

This paper presents a multi-pipeline Japanese-to-English machine translation (MT) system designed to address the challenge of translating repeated elements from Japanese into fluent and lexically diverse English. The system is developed as part of the Non-Repetitive Translation Task at WMT24, which focuses on minimizing redundancy while maintaining high translation quality. Our approach utilizes MeCab, the de facto NLP tool for Japanese, for the identification of repeated elements, and Claude Sonnet 3.5, a large language model (LLM), for translation and proofreading. The system effectively accomplishes the shared task by identifying and translating in a diversified manner 89.79% of the 470 repeated instances in the testing dataset, and achieving an average translation quality score of 4.60 out of 5, significantly surpassing the baseline score of 3.88. Analysis also revealed the challenges encountered, particularly in identifying standalone noun-suffix elements and occasional cases of consistent translations or mistranslations.

This paper describes the results of the first shared task on generation of teacher responses in educational dialogues. The goal of the task was to benchmark the ability of generative language models to act as AI teachers, replying to a student in a teacher-student dialogue. Eight teams participated in the competition hosted on CodaLab and experimented with a wide variety of state-of-the-art models, including Alpaca, Bloom, DialoGPT, DistilGPT-2, Flan-T5, GPT- 2, GPT-3, GPT-4, LLaMA, OPT-2.7B, and T5- base. Their submissions were automatically scored using BERTScore and DialogRPT metrics, and the top three among them were further manually evaluated in terms of pedagogical ability based on Tack and Piech (2022). The NAISTeacher system, which ranked first in both automated and human evaluation, generated responses with GPT-3.5 Turbo using an ensemble of prompts and DialogRPT-based ranking of responses for given dialogue contexts. Despite promising achievements of the participating teams, the results also highlight the need for evaluation metrics better suited to educational contexts.

Grammatical Error Correction (GEC) is the task of automatically detecting and correcting errors in text. The task not only includes the correction of grammatical errors, such as missing prepositions and mismatched subject–verb agreement, but also orthographic and semantic errors, such as misspellings and word choice errors, respectively. The field has seen significant progress in the last decade, motivated in part by a series of five shared tasks, which drove the development of rule-based methods, statistical classifiers, statistical machine translation, and finally neural machine translation systems, which represent the current dominant state of the art. In this survey paper, we condense the field into a single article and first outline some of the linguistic challenges of the task, introduce the most popular datasets that are available to researchers (for both English and other languages), and summarize the various methods and techniques that have been developed with a particular focus on artificial error generation. We next describe the many different approaches to evaluation as well as concerns surrounding metric reliability, especially in relation to subjective human judgments, before concluding with an overview of recent progress and suggestions for future work and remaining challenges. We hope that this survey will serve as a comprehensive resource for researchers who are new to the field or who want to be kept apprised of recent developments.

Large Language Models (LLMs) evaluation is a patchy and inconsistent landscape, and it is becoming clear that the quality of automatic evaluation metrics is not keeping up with the pace of development of generative models. We aim to improve the understanding of current models’ performance by providing a preliminary and hybrid evaluation on a range of open and closed-source generative LLMs on three NLP benchmarks: text summarisation, text simplification and grammatical error correction (GEC), using both automatic and human evaluation. We also explore the potential of the recently released GPT-4 to act as an evaluator. We find that ChatGPT consistently outperforms many other popular models according to human reviewers on the majority of metrics, while scoring much more poorly when using classic automatic evaluation metrics. We also find that human reviewers rate the gold reference as much worse than the best models’ outputs, indicating the poor quality of many popular benchmarks. Finally, we find that GPT-4 is capable of ranking models’ outputs in a way which aligns reasonably closely to human judgement despite task-specific variations, with a lower alignment in the GEC task.

We extend a current sequence-tagging approach to Grammatical Error Correction (GEC) by introducing specialised tags for spelling correction and morphological inflection using the SymSpell and LemmInflect algorithms. Our approach improves generalisation: the proposed new tagset allows a smaller number of tags to correct a larger range of errors. Our results show a performance improvement both overall and in the targeted error categories. We further show that ensembles trained with our new tagset outperform those trained with the baseline tagset on the public BEA benchmark.

This paper presents an investigation into the differences between processing monolingual input and code-switching (CSW) input in the context of machine translation (MT). Specifically, we compare the performance of three MT systems (Google, mBART-50 and M2M-100-big) in terms of their ability to translate monolingual Vietnamese, a low-resource language, and Vietnamese-English CSW respectively. To our knowledge, this is the first study to systematically analyse what might happen when multilingual MT systems are exposed to CSW data using both automatic and human metrics. We find that state-of-the-art neural translation systems not only achieve higher scores on automatic metrics when processing CSW input (compared to monolingual input), but also produce translations that are consistently rated as more semantically faithful by humans. We further suggest that automatic evaluation alone is insufficient for evaluating the translation of CSW input. Our findings establish a new benchmark that offers insights into the relationship between MT and CSW.

Grammatical error correction (GEC) is the task of automatically correcting errors in text. It has mainly been developed to assist language learning, but can also be applied to native text. This paper reports on preliminary work in improving GEC for multiword expression (MWE) error correction. We propose two systems which incorporate MWE information in two different ways: one is a multi-encoder decoder system which encodes MWE tags in a second encoder, and the other is a BART pre-trained transformer-based system that encodes MWE representations using special tokens. We show improvements in correcting specific types of verbal MWEs based on a modified version of a standard GEC evaluation approach.

Document-level context can provide valuable information in grammatical error correction (GEC), which is crucial for correcting certain errors and resolving inconsistencies. In this paper, we investigate context-aware approaches and propose document-level GEC systems. Additionally, we employ a three-step training strategy to benefit from both sentence-level and document-level data. Our system outperforms previous document-level and all other NMT-based single-model systems, achieving state of the art on a common test set.

In this paper, we show how a multi-class grammatical error detection (GED) system can be used to improve grammatical error correction (GEC) for English. Specifically, we first develop a new state-of-the-art binary detection system based on pre-trained ELECTRA, and then extend it to multi-class detection using different error type tagsets derived from the ERRANT framework. Output from this detection system is used as auxiliary input to fine-tune a novel encoder-decoder GEC model, and we subsequently re-rank the N-best GEC output to find the hypothesis that most agrees with the GED output. Results show that fine-tuning the GEC system using 4-class GED produces the best model, but re-ranking using 55-class GED leads to the best performance overall. This suggests that different multi-class GED systems benefit GEC in different ways. Ultimately, our system outperforms all other previous work that combines GED and GEC, and achieves a new single-model NMT-based state of the art on the BEA-test benchmark.

This paper describes our submission to the SemEval-2021 shared task on Lexical Complexity Prediction. We approached it as a regression problem and present an ensemble combining four systems, one feature-based and three neural with fine-tuning, frequency pre-training and multi-task learning, achieving Pearson scores of 0.8264 and 0.7556 on the trial and test sets respectively (sub-task 1). We further present our analysis of the results and discuss our findings.

This paper describes the system of the Cambridge team submitted to the SemEval-2021 shared task on Multilingual and Cross-lingual Word-in-Context Disambiguation. Building on top of a pre-trained masked language model, our system is first pre-trained on out-of-domain data, and then fine-tuned on in-domain data. We demonstrate the effectiveness of the proposed two-step training strategy and the benefits of data augmentation from both existing examples and new resources. We further investigate different representations and show that the addition of distance-based features is helpful in the word-in-context disambiguation task. Our system yields highly competitive results in the cross-lingual track without training on any cross-lingual data; and achieves state-of-the-art results in the multilingual track, ranking first in two languages (Arabic and Russian) and second in French out of 171 submitted systems.

In this paper, we describe our submission to the BEA 2019 shared task on grammatical error correction. We present a system pipeline that utilises both error detection and correction models. The input text is first corrected by two complementary neural machine translation systems: one using convolutional networks and multi-task learning, and another using a neural Transformer-based system. Training is performed on publicly available data, along with artificial examples generated through back-translation. The n-best lists of these two machine translation systems are then combined and scored using a finite state transducer (FST). Finally, an unsupervised re-ranking system is applied to the n-best output of the FST. The re-ranker uses a number of error detection features to re-rank the FST n-best list and identify the final 1-best correction hypothesis. Our system achieves 66.75% F 0.5 on error correction (ranking 4th), and 82.52% F 0.5 on token-level error detection (ranking 2nd) in the restricted track of the shared task.

This paper describes our use of two recurrent neural network sequence models: sequence labelling and sequence-to-sequence models, for the prediction of future learner errors in our submission to the 2018 Duolingo Shared Task on Second Language Acquisition Modeling (SLAM). We show that these two models capture complementary information as combining them improves performance. Furthermore, the same network architecture and group of features can be used directly to build competitive prediction models in all three language tracks, demonstrating that our approach generalises well across languages.

We propose an approach to N-best list reranking using neural sequence-labelling models. We train a compositional model for error detection that calculates the probability of each token in a sentence being correct or incorrect, utilising the full sentence as context. Using the error detection model, we then re-rank the N best hypotheses generated by statistical machine translation systems. Our approach achieves state-of-the-art results on error correction for three different datasets, and it has the additional advantage of only using a small set of easily computed features that require no linguistic input.

Shortage of available training data is holding back progress in the area of automated error detection. This paper investigates two alternative methods for artificially generating writing errors, in order to create additional resources. We propose treating error generation as a machine translation task, where grammatically correct text is translated to contain errors. In addition, we explore a system for extracting textual patterns from an annotated corpus, which can then be used to insert errors into grammatically correct sentences. Our experiments show that the inclusion of artificially generated errors significantly improves error detection accuracy on both FCE and CoNLL 2014 datasets.