Papers by Alexander Mehler
LTV: Labeled Topic Vector (C18-2)
Copied to clipboard
| Challenge: | Using nnDDC, we generate labeled topic classifications based on the Dewey Decimal Classification (DDC) Unlike related approaches, we use classifiers to define the dimensions of CISS, which are directly labeles by the underlying target class. |
| Approach: | They propose a website and API that generates labeled topic classifications based on the Dewey Decimal Classification (DDC) they propose nnDDC, a largely language-independent natural network-based classifier for DDC, which is language-dependent . |
| Outcome: | The proposed model is language-independent and performs well in 40 languages. |
Recognizing Sentence-level Logical Document Structures with the Help of Context-free Grammars (2020.lrec-1)
Copied to clipboard
| Challenge: | Current sentence boundary detectors split documents into sequentially ordered sentences without their dependencies. |
| Approach: | They propose a tool that segments sentences into tree structures to detect recursive structure . they retrain different constituency parsers to transform them into sentence segmenters . |
| Outcome: | The proposed tool can detect recursive structure in documents with a main clause and subordinate clauses . the proposed tool improves German dependency parsing by providing additional structural information. |
When Specialization Helps: Using Pooled Contextualized Embeddings to Detect Chemical and Biomedical Entities in Spanish (D19-57)
Copied to clipboard
| Challenge: | Existing work on pharmacological entities requires manual annotation of these units. |
| Approach: | They propose an approach to task 1 of the PharmaCoNER Challenge to recognize pharmacological entities on a spanish corpus. |
| Outcome: | The proposed approach achieves 89.76% score on a spanish corpus based on pre-trained embeddings and 90.52% score on domain-specific embeddables. |
Filling the Temporal Void: Recovering Missing Publication Years in the Project Gutenberg Corpus Using LLMs (2025.findings-acl)
Copied to clipboard
| Challenge: | Currently, there is no publicly available corpus for diachronic text analysis due to the lack of accurate temporal metadata. |
| Approach: | They propose to add missing temporal metadata to the Gutenberg corpus by using open web, Wikipedia, and Open Library API sources. |
| Outcome: | The proposed corpus includes 53,774 books with a total of 3.8 billion tokens in 11 languages, produced between 1600 and 2000. |
I still have Time(s): Extending HeidelTime for German Texts (2022.lrec-1)
Copied to clipboard
| Challenge: | HeidelTime is a widely used tool for detecting temporal expressions in texts. |
| Approach: | They propose to extend HeidelTime's pattern matching system by observing false negatives within real world texts and various time banks. |
| Outcome: | The proposed extension can detect expressions in texts and time banks in a convenient way. |
On the Influence of Coreference Resolution on Word Embeddings in Lexical-semantic Evaluation Tasks (2020.lrec-1)
Copied to clipboard
| Challenge: | Existing word embeddings rely on local information delimited by context windows or dependency parents to predict word relations. |
| Approach: | They propose to use coreference resolution to find all spans of a text that refer to the same entity to improve the F1-Scores. |
| Outcome: | The proposed methods do not benefit significantly from pronoun substitution. |
Tafsir Dataset: A Novel Multi-Task Benchmark for Named Entity Recognition and Topic Modeling in Classical Arabic Literature (2022.coling-1)
Copied to clipboard
Sajawel Ahmed, Rob van der Goot, Misbahur Rehman, Carl Kruse, Ömer Özsoy, Alexander Mehler, Gemma Roig
| Challenge: | Named entity recognition and topic modeling are crucial for downstream tasks in natural language processing. |
| Approach: | They propose to address named entity recognition and topic modeling on CA literature . they manually annotate the work of Tafsir Al-Tabari with span-based NEs . |
| Outcome: | The results show that the proposed task can perform state-of-the-art on historical topic models. |
FastSense: An Efficient Word Sense Disambiguation Classifier (L18-1)
Copied to clipboard
| Challenge: | Word Sense Disambiguation (WSD) is a task that is often overlooked by NLP pipelines because of its complexity and complexity. |
| Approach: | They propose a neural network-based tool for word sense disambiguation called fastSense. |
| Outcome: | The proposed tool can process huge amounts of data quickly and surpasses state-of-the-art tools in terms of F-measure. |
German Parliamentary Corpus (GerParCor) (2022.lrec-1)
Copied to clipboard
| Challenge: | German parliaments have a large and partly unexploited treasure trove of publicly accessible texts. |
| Approach: | a new corpus of German-language parliamentary protocols is made available in XMI format . the corpus is genre-specific and contains conversions of scanned protocols . a researcher at the university of berlin and a professor at the berlin university created the corpuus . |
| Outcome: | the German Parliamentary Corpus is a genre-specific corpus of German-language parliamentary protocols from three centuries and four countries. |
MedLinkDE – MedDRA Entity Linking for German with Guided Chain of Thought Reasoning (2025.emnlp-main)
Copied to clipboard
Roman Christof, Farnaz Zeidi, Manuela Messelhäußer, Dirk Mentzer, Renate Koenig, Liam Childs, Alexander Mehler
| Challenge: | In pharmacovigilance, effective automation of medical data structuring is critical. |
| Approach: | They propose a German real-world MedDRA dataset that uses embedding models to retrieve medical terms and guided chain-of-thought re-ranking using LLMs. |
| Outcome: | The proposed approach shows that it is feasible to integrate German data with German MedDRA datasets under the constraints of data scarcity. |
German Parliamentary Corpus (GerParCor) Reloaded (2024.lrec-main)
Copied to clipboard
| Challenge: | In 2022, the largest German-speaking corpus of parliamentary protocols from three different centuries has been published - GerParCor. |
| Approach: | They propose to update the largest German-speaking corpus of parliamentary protocols from three different centuries, on a national and federal level, from Germany, Austria, Switzerland and Liechtenstein, and to make them available in XMI format. |
| Outcome: | The updated corpus includes all new parliamentary protocols and adds and preprocesses further parliamentary protocol not covered in the previous version. |
A UIMA Database Interface for Managing NLP-related Text Annotations (L18-1)
Copied to clipboard
| Challenge: | despite the use of UIMA as a document-based schema, it does not provide native database support. |
| Approach: | They develop a database interface to allow generic use of UIMA documents in database systems. |
| Outcome: | The framework is evaluated in relation to file system-based storage and provides data protection. |
TextAnnotator: A UIMA Based Tool for the Simultaneous and Collaborative Annotation of Texts (2020.lrec-1)
Copied to clipboard
| Challenge: | Existing annotation tools are not efficient for the annotation of corpora and are not error-free. |
| Approach: | They propose to extend existing annotation tools by evaluating their flexibility and efficiency. |
| Outcome: | The proposed system performs platform-independent multimodal annotations and annotates complex textual structures. |
What do Toothbrushes do in the Kitchen? How Transformers Think our World is Structured (2022.naacl-main)
Copied to clipboard
| Challenge: | Recent research reveals that transformer-based models are biased towards extracting knowledge about object relations. |
| Approach: | They propose to use transformer-based models to extract knowledge about object relations to investigate whether they can be used to extract object relations. |
| Outcome: | The proposed models outperform static models in many respects and perform much worse than similarity measures and classifiers. |
Unlocking the Heterogeneous Landscape of Big Data NLP with DUUI (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Automated analysis of large corpora is a complex task, especially in terms of time efficiency. |
| Approach: | They propose a framework for automatic distributed analysis of text corpora that leverages Big Data experience and virtualization with Docker. |
| Outcome: | The proposed framework is scalable, flexible, lightweight, and feature-rich for automatic distributed analysis of text corpora. |
WikiDragon: A Java Framework For Diachronic Content And Network Analysis Of MediaWikis (L18-1)
Copied to clipboard
| Challenge: | WikiDragon is a Java Framework designed to give developers in computational linguistics an intuitive API to build, parse and analyze instances of MediaWikis. |
| Approach: | They introduce WikiDragon, a Java Framework that allows developers to build, parse and analyze instances of MediaWikis on their computers. |
| Outcome: | The framework is based on the Wikipedia, Wiktionary, WikiSource or WikiNews and evaluates link extraction, diachronic network analysis and the impact of different frameworks to text analysis. |
German SRL: Corpus Construction and Model Training (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing semantic role annotation resources are lacking for German. |
| Approach: | They propose a translation-based approach to train German semantic role models using semantic annotations and alignment models. |
| Outcome: | The proposed method achieves competitive evaluation scores, but avoids limitations of previous approaches. |
Dependencies over Times and Tools (DoTT) (2024.lrec-main)
Copied to clipboard
Andy Luecking, Giuseppe Abrami, Leon Hammerla, Marc Rahn, Daniel Baumartz, Steffen Eger, Alexander Mehler
| Challenge: | Using the examples of English and German, we examine how parsers trained on modern variants of these languages can be transferred to older language levels without loss. |
| Approach: | They develop a treebank of diachronic corpora enriched with dependency annotations using 3 parsers, 6 pre-trained language models, 5 newly trained models for German, and two tag sets. |
| Outcome: | The proposed treebank covers the time period from 1800 until today and is based on the DependencyAnnotator annotation tool. |
TreeAnnotator: Versatile Visual Annotation of Hierarchical Text Relations (L18-1)
Copied to clipboard
| Challenge: | TREEANNOTATOR is a browser-based tool for annotating tree-like structures . it provides a wider range of formats and provides graphical annotations . |
| Approach: | They evaluate TREEANNOTATOR, a browser-based tool for annotating tree-like structures, in particular structures that jointly map dependency relations and inclusion hierarchies, as used by Rhetorical Structure Theory. |
| Outcome: | The GUI interface is user-friendly and provides two visualization modes. |