Jelena Mitrović

Abdullah Al Zubaer, Lorenz Wendlinger, Simon Alexander Nonn et al.

Grading German legal exam solutions faces growing volumes and a shortage of qualified graders, delaying feedback and creating a bottleneck. At the same time, it is a high-stakes expert task, since state exam grades strongly influence career outcomes in Germany. Despite this practical relevance, literature lacks systematic studies on effective methods for grading legal exams. To address this gap, we investigate whether large language models (LLMs) can support the automated grading of German legal case solutions in criminal and public law, thereby enabling scalable feedback and student self-testing. We present a systematic evaluation of 27 proprietary and open-source LLMs, benchmarking prompting strategies that incrementally add task-related information, such as a sample solution and a grading rubric. Using quadratic weighted kappa (QWK), reasoning-oriented LLMs can approximate expert grading in public law when given a sample solution and a grading rubric (up to 0.91), compared to 0.60 in criminal law, suggesting a harder grading task in criminal law. Beyond single-model grading, ensembling improves agreement by up to 0.15 over its best member and can offer an alternative to stronger closed-source single models. In addition, our findings suggest that effective prompt design and model selection are necessary for reliable LLM-based grading of legal exams.

Timo Spinde, Jan-David Krieger, Terry Ruas et al.

Media has a substantial impact on the public perception of events. A one-sided or polarizing perspective on any topic is usually described as media bias. One of the ways how bias in news articles can be introduced is by altering word choice. Biased word choices are not always obvious, nor do they exhibit high context-dependency. Hence, detecting bias is often difficult. We propose a Transformer-based deep learning architecture trained via Multi-Task Learning using six bias-related data sets to tackle the media bias detection problem. Our best-performing implementation achieves a macro $F_{1}$ of 0.776, a performance boost of 3\% compared to our baseline, outperforming existing methods. Our results indicate Multi-Task Learning as a promising alternative to improve existing baseline models in identifying slanted reporting.

Harshil Darji, Jelena Mitrović, Michael Granitzer

The use of BERT, one of the most popular language models, has led to improvements in many Natural Language Processing (NLP) tasks. One such task is Named Entity Recognition (NER) i.e. automatic identification of named entities such as location, person, organization, etc. from a given text. It is also an important base step for many NLP tasks such as information extraction and argumentation mining. Even though there is much research done on NER using BERT and other popular language models, the same is not explored in detail when it comes to Legal NLP or Legal Tech. Legal NLP applies various NLP techniques such as sentence similarity or NER specifically on legal data. There are only a handful of models for NER tasks using BERT language models, however, none of these are aimed at legal documents in German. In this paper, we fine-tune a popular BERT language model trained on German data (German BERT) on a Legal Entity Recognition (LER) dataset. To make sure our model is not overfitting, we performed a stratified 10-fold cross-validation. The results we achieve by fine-tuning German BERT on the LER dataset outperform the BiLSTM-CRF+ model used by the authors of the same LER dataset. Finally, we make the model openly available via HuggingFace.

Mehdi Ben Amor, Michael Granitzer, Jelena Mitrović

Language Models (LMs) have shown state-of-the-art performance in Natural Language Processing (NLP) tasks. Downstream tasks such as Named Entity Recognition (NER) or Part-of-Speech (POS) tagging are known to suffer from data imbalance issues, particularly regarding the ratio of positive to negative examples and class disparities. This paper investigates an often-overlooked issue of encoder models, specifically the position bias of positive examples in token classification tasks. For completeness, we also include decoders in the evaluation. We evaluate the impact of position bias using different position embedding techniques, focusing on BERT with Absolute Position Embedding (APE), Relative Position Embedding (RPE), and Rotary Position Embedding (RoPE). Therefore, we conduct an in-depth evaluation of the impact of position bias on the performance of LMs when fine-tuned on token classification benchmarks. Our study includes CoNLL03 and OntoNote5.0 for NER, English Tree Bank UD\_en, and TweeBank for POS tagging. We propose an evaluation approach to investigate position bias in transformer models. We show that LMs can suffer from this bias with an average drop ranging from 3\% to 9\% in their performance. To mitigate this effect, we propose two methods: Random Position Shifting and Context Perturbation, that we apply on batches during the training process. The results show an improvement of $\approx$ 2\% in the performance of the model on CoNLL03, UD\_en, and TweeBank.

Christofer Fellicious, Lorenz Wendlinger, Mario Gancarski et al.

Supervised machine learning often encounters concept drift, where the data distribution changes over time, degrading model performance. Existing drift detection methods focus on identifying these shifts but often overlook the challenge of acquiring labeled data for model retraining after a shift occurs. We present the Strategy for Drift Sampling (SUDS), a novel method that selects homogeneous samples for retraining using existing drift detection algorithms, thereby enhancing model adaptability to evolving data. SUDS seamlessly integrates with current drift detection techniques. We also introduce the Harmonized Annotated Data Accuracy Metric (HADAM), a metric that evaluates classifier performance in relation to the quantity of annotated data required to achieve the stated performance, thereby taking into account the difficulty of acquiring labeled data. Our contributions are twofold: SUDS combines drift detection with strategic sampling to improve the retraining process, and HADAM provides a metric that balances classifier performance with the amount of labeled data, ensuring efficient resource utilization. Empirical results demonstrate the efficacy of SUDS in optimizing labeled data use in dynamic environments, significantly improving the performance of machine learning applications in real-world scenarios. Our code is open source and available at https://github.com/cfellicious/SUDS/

Ramona Kühn, Jelena Mitrović, Michael Granitzer

Rhetorical figures play an important role in our communication. They are used to convey subtle, implicit meaning, or to emphasize statements. We notice them in hate speech, fake news, and propaganda. By improving the systems for computational detection of rhetorical figures, we can also improve tasks such as hate speech and fake news detection, sentiment analysis, opinion mining, or argument mining. Unfortunately, there is a lack of annotated data, as well as qualified annotators that would help us build large corpora to train machine learning models for the detection of rhetorical figures. The situation is particularly difficult in languages other than English, and for rhetorical figures other than metaphor, sarcasm, and irony. To overcome this issue, we develop a web application called "Find your Figure" that facilitates the identification and annotation of German rhetorical figures. The application is based on the German Rhetorical ontology GRhOOT which we have specially adapted for this purpose. In addition, we improve the user experience with Retrieval Augmented Generation (RAG). In this paper, we present the restructuring of the ontology, the development of the web application, and the built-in RAG pipeline. We also identify the optimal RAG settings for our application. Our approach is one of the first to practically use rhetorical ontologies in combination with RAG and shows promising results.

Mohamed Ayoub Ben Ayad, Jelena Mitrovic, Michael Granitzer

We introduce Krony-PT, a compression technique for GPT-2 based on Kronecker products. We specifically target the feed-forward weights of each transformer block, and systematically compress the feed-forward layer matrices to various degrees. We introduce a modified Van Loan decomposition to initialize new Kronecker factors, and also propose a new pruning-based initialization technique. Our method compresses the original 124M-parameter GPT-2 to various smaller models, ranging from 80M to 96M. Our 81M model variant outperforms DistilGPT2 on next-token prediction across all standard language modeling datasets, and shows competitive or comparable performance with significantly larger Kronecker-based compressions of GPT-2.

Mohamed Ayoub Ben Ayad, Michael Dinzinger, Kanishka Ghosh Dastidar et al.

Embedding models are central to dense retrieval, semantic search, and recommendation systems, but their size often makes them impractical to deploy in resource-constrained environments such as browsers or edge devices. While smaller embedding models offer practical advantages, they typically underperform compared to their larger counterparts. To bridge this gap, we demonstrate that concatenating the raw embedding vectors of multiple small models can outperform a single larger baseline on standard retrieval benchmarks. To overcome the resulting high dimensionality of naive concatenation, we introduce a lightweight unified decoder trained with a Matryoshka Representation Learning (MRL) loss. This decoder maps the high-dimensional joint representation to a low-dimensional space, preserving most of the original performance without fine-tuning the base models. We also show that while concatenating more base models yields diminishing gains, the robustness of the decoder's representation under compression and quantization improves. Our experiments show that, on a subset of MTEB retrieval tasks, our concat-encode-quantize pipeline recovers 89\% of the original performance with a 48x compression factor when the pipeline is applied to a concatenation of four small embedding models.

Ramona Kühn, Jelena Mitrović, Michael Granitzer

Rhetorical figures play a major role in our everyday communication as they make text more interesting, more memorable, or more persuasive. Therefore, it is important to computationally detect rhetorical figures to fully understand the meaning of a text. We provide a comprehensive overview of computational approaches to lesser-known rhetorical figures. We explore the linguistic and computational perspectives on rhetorical figures, emphasizing their significance for the domain of Natural Language Processing. We present different figures in detail, delving into datasets, definitions, rhetorical functions, and detection approaches. We identified challenges such as dataset scarcity, language limitations, and reliance on rule-based methods.

Tommaso Caselli, Valerio Basile, Jelena Mitrović et al.

In this paper, we introduce HateBERT, a re-trained BERT model for abusive language detection in English. The model was trained on RAL-E, a large-scale dataset of Reddit comments in English from communities banned for being offensive, abusive, or hateful that we have collected and made available to the public. We present the results of a detailed comparison between a general pre-trained language model and the abuse-inclined version obtained by retraining with posts from the banned communities on three English datasets for offensive, abusive language and hate speech detection tasks. In all datasets, HateBERT outperforms the corresponding general BERT model. We also discuss a battery of experiments comparing the portability of the generic pre-trained language model and its corresponding abusive language-inclined counterpart across the datasets, indicating that portability is affected by compatibility of the annotated phenomena.