Handwriting remains an essential skill, particularly in education. Therefore, providing visual feedback on handwritten documents is an important but understudied area. We outline the many challenges when going from an image of handwritten input to correctly placed informative error feedback. We empirically compare modular and end-to-end systems and find that both approaches currently do not achieve acceptable overall quality. We identify the major challenges and outline an agenda for future research.

This paper introduces MultiGEC, a dataset for multilingual Grammatical Error Correction (GEC) in twelve European languages: Czech, English, Estonian, German, Greek, Icelandic, Italian, Latvian, Russian, Slovene, Swedish and Ukrainian. MultiGEC distinguishes itself from previous GEC datasets in that it covers several underrepresented languages, which we argue should be included in resources used to train models for Natural Language Processing tasks which, as GEC itself, have implications for Learner Corpus Research and Second Language Acquisition. Aside from multilingualism, the novelty of the MultiGEC dataset is that it consists of full texts — typically learner essays — rather than individual sentences, making it possible to train systems that take a broader context into account. The dataset was built for MultiGEC-2025, the first shared task in multilingual text-level GEC, but it remains accessible after its competitive phase, serving as a resource to train new error correction systems and perform cross-lingual GEC studies.

In this work, we assess the potential of using synthetic data to train models for content scoring. We generate a parallel corpus of LLM-generated data for the SRA dataset. In our experiments, we train three different kinds of models (Logistic Regression, BERT, SBERT) with this data, examining their respective ability to bridge between generated training data and student-authored test data. We also explore the effects of generating larger volumes of training data than what is available in the original dataset. Overall, we find that training models from LLM-generated data outperforms zero-shot scoring of the test data with an LLM. Still, the fine-tuned models perform much worse than models trained on the original data, largely because the LLM-generated answers often do not to conform to the desired labels. However, once the data is manually relabeled, competitive models can be trained from it. With a similarity-based scoring approach, the relabeled (larger) amount of synthetic answers consistently yields a model that surpasses performance of training on the (limited) amount of answers available in the original dataset.

We present the framework Omethi, which is aimed at scoring short text responses in a semi-automatic fashion, particularly fit to international large-scale assessments. We evaluate its effectiveness for the massively multilingual PISA tests. Responses are passed through a conditional flow of hierarchically combined scoring components to assign a score. Once a score is assigned, hierarchically lower components are discarded. Models implemented in this study ranged from lexical matching of normalized texts-with excellent accuracy but weak generalizability-to fine-tuned large language models-with lower accuracy but high generalizability. If not scored by any automatic component, responses are passed on to manual scoring. The paper is the first to provide an evaluation of automatic scoring on multilingual PISA data in eleven languages (including Arabic, Finnish, Hebrew, and Kazakh) from three domains (n = 3.8 million responses). On average, results show a manual effort reduction of 71 percent alongside an agreement of kappa = .957, when including manual scoring, and kappa = .804 for only the automatically scored responses. The evaluation underscores the framework's effective adaptivity and operational feasibility with its shares of used components varying substantially across domains and languages while maintaining homogeneously high accuracy.

The assessment of legal problems requires the consideration of a specific legal system and its levels of abstraction, from constitutional law to statutory law to case law. The extent to which Large Language Models (LLMs) internalize such legal systems is unknown. In this paper, we propose and investigate different approaches to condition LLMs at different levels of abstraction in legal systems. This paper examines different approaches to conditioning LLMs at multiple levels of abstraction in legal systems to detect potentially punishable hate speech. We focus on the task of classifying whether a specific social media posts falls under the criminal offense of incitement to hatred as prescribed by the German Criminal Code. The results show that there is still a significant performance gap between models and legal experts in the legal assessment of hate speech, regardless of the level of abstraction with which the models were conditioned. Our analysis revealed, that models conditioned on abstract legal knowledge lacked deep task understanding, often contradicting themselves and hallucinating answers, while models using concrete legal knowledge performed reasonably well in identifying relevant target groups, but struggled with classifying target conducts.

Real-word spelling errors (RWSEs) pose special challenges for detection methods, as they 'hide' in the form of another existing word and in many cases even fit in syntactically. We present a modern Transformer-based implementation of earlier probabilistic methods based on confusion sets and show that RWSEs can be detected with a good balance between missing errors and raising too many false alarms. The confusion sets are dynamically configurable, allowing teachers to easily adjust which errors trigger feedback.

The evaluation of learning and assessment content has always been a crucial task in the educational domain, but traditional approaches based on human feedback are not always usable in modern educational settings. Indeed, the advent of machine learning models, in particular Large Language Models (LLMs), enabled to quickly and automatically generate large quantities of texts, making human evaluation unfeasible. Still, these texts are used in the educational domain – e.g., as questions, hints, or even to score and assess students – and thus the need for accurate and automated techniques for evaluation becomes pressing. This workshop aims to attract professionals from both academia and the industry, and to to offer an opportunity to discuss which are the common challenges in evaluating learning and assessment content in education.

With the recent emergence of powerful visiolinguistic models comes the question of how fine-grained their multi-modal understanding is. This has lead to the release of several probing datasets. Results point towards models having trouble with prepositions and verbs, but being relatively robust when it comes to color. To gauge how deep this understanding goes, we compile a comprehensive probing dataset to systematically test multi-modal alignment around color. We demonstrate how human perception influences descriptions of color and pay special attention to the extent to which this is reflected within the predictions of a visiolinguistic model. Probing a set of models with diverse properties with our benchmark confirms the superiority of models that do not rely on pre-extracted image features, and demonstrates that augmentation with too much noisy pre-training data can produce an inferior model. While the benchmark remains challenging for all models we test, the overall result pattern suggests well-founded alignment of color terms with hues. Analyses do however reveal uncertainty regarding the boundaries between neighboring color terms.

Despite advances in machine learning based hate speech detection, the need for larges amounts of labeled training data for state-of-the-art approaches remains a challenge for their application. Semi-supervised learning addresses this problem by leveraging unlabeled data and thus reducing the amount of annotated data required. Underlying this approach is the assumption that labeled and unlabeled data follow similar distributions. This assumption however may not always hold, with consequences for real world applications. We address this problem by investigating the dynamics of pseudo-labeling, a commonly employed form of semi-supervised learning, in the context of hate speech detection. Concretely we analysed the influence of data characteristics and of two strategies for selecting pseudo-labeled samples: threshold- and ratio-based. The results show that the influence of data characteristics on the pseudo-labeling performances depends on other factors, such as pseudo-label selection strategies or model biases. Furthermore, the effectiveness of pseudo-labeling in classification performance is determined by the interaction between the number, hate ratio and accuracy of the selected pseudo-labels. Analysis of the results suggests an advantage of the threshold-based approach when labeled and unlabeled data arise from the same domain, whilst the ratio-based approach may be recommended in the opposite situation.

A possible way to save manual grading effort in short answer scoring is to automatically score answers for which the classifier is highly confident. We explore the feasibility of this approach in a high-stakes exam setting, evaluating three different similarity-based scoring methods, where the similarity score is a direct proxy for model confidence. The decision on an appropriate level of confidence should ideally be made before scoring a new prompt. We thus probe to what extent confidence thresholds are consistent across different datasets and prompts. We find that high-confidence thresholds vary on a prompt-to-prompt basis, and that the overall potential of increased performance at a reasonable cost of additional manual effort is limited.

Using the full potential of LLMs in higher education is hindered by challenges with access to LLMs. The two main access modes currently discussed are paying for a cloud-based LLM or providing a locally maintained open LLM. In this paper, we describe the current state of establishing an open LLM infrastructure at FernUniversität in Hagen under the project name FLEXI (FernUni LLM Experimental Infrastructure). FLEXI enables experimentation within teaching and research with the goal of generating strongly needed evidence in favor (or against) the use of locally maintained open LLMs in higher education. The paper will provide some practical guidance for everyone trying to decide whether to run their own LLM server.

Recent advances in natural language processing (NLP) can be largely attributed to the advent of pre-trained language models such as BERT and RoBERTa. While these models demonstrate remarkable performance on general datasets, they can struggle in specialized domains such as medicine, where unique domain-specific terminologies, domain-specific abbreviations, and varying document structures are common. This paper explores strategies for adapting these models to domain-specific requirements, primarily through continuous pre-training on domain-specific data. We pre-trained several German medical language models on 2.4B tokens derived from translated public English medical data and 3B tokens of German clinical data. The resulting models were evaluated on various German downstream tasks, including named entity recognition (NER), multi-label classification, and extractive question answering. Our results suggest that models augmented by clinical and translation-based pre-training typically outperform general domain models in medical contexts. We conclude that continuous pre-training has demonstrated the ability to match or even exceed the performance of clinical models trained from scratch. Furthermore, pre-training on clinical data or leveraging translated texts have proven to be reliable methods for domain adaptation in medical NLP tasks.

Using the full potential of LLMs in higher education is hindered by challenges with access to LLMs. The two main access modes currently discussed are paying for a cloud-based LLM or providing a locally maintained open LLM. In this paper, we describe the current state of establishing an open LLM infrastructure at FernUniversiat in Hagen (Germany) under the project name FLEXI (Flexible LLM Experimental Infrastructure). FLEXI enables experimentation within teaching and research with the goal of generating strongly needed evidence in favor (or against) the use of locally maintained open LLMs in higher education. The paper will provide some practical guidance for everyone

Free-text tasks, where students need to write a short answer to a specific question, serve as a well-established method for assessing learner knowledge. To address the high cost of manually scoring these tasks, automated scoring models can be used. Such models come in various types, each with its own strengths and weaknesses. Comparing these models helps in selecting the most suitable one for a given problem. Depending on the assessment context, this decision can be driven by ethical or legal considerations. When implemented successfully, a scoring model has the potential to substantially reduce costs and enhance the reliability of the scoring process. This article compares the different categories of scoring models across a set of crucial criteria that have immediate relevance to model employment in practice.

This study investigates the potential of Large Language Models (LLMs), in particular GPT and LLaMA, for automated scoring of short answer responses. We focus on zero-shot and few-shot settings, but also compare with fine-tuned models and a supervised upper-bound. Our results show that LLMs perform much worse in those settings on a performance level that is not feasible for practical purposes. Fine-tuning LLMs brings their results on roughly the same level as supervised results, but as they are less efficient there currently seems to be no basis for applying LLMs for short answer scoring.