Human Language Technology @ Augsburg

News Archive

Archive of the regular updates from our groups.

February 21, 2024 -- Our paper AnnoCTR: A Dataset for Detecting and Linking Entities, Tactics, and Techniques in Cyber Threat Reports has been accepted to LREC/Coling, the 2024 Joint International Conference on Computational Linguistics, Language Resources, and Evaluation, which will take place in May in Torino, Italy. In the paper, which is a collaboration with Bosch and the MPI of Informatics in Saarbrücken, we present a new dataset of classifying cyber attacks and techniques in reports. Also, the paper Explaining Pre-Trained Language Models with Attribution Scores: An Analysis in Low-Resource Settings by our Phd student Wei Zhou, together with Bosch and Univerisity of Stuttgart, has been accepted at LREC-COLING 2024.
February 8, 2024 -- Annemarie Friedrich has been elected as President of the German Society for Computational Linguistics and Language Technology (GSCL).
January 24, 2024 -- Our paper Aligning Uncertainty: Leveraging LLMs to Analyze Uncertainty Transfer in Text Summarization has been accepted to the UncertaiNLP Workshop at EACL 2024. Congrats, Zahra!
January 17, 2024 -- Annemarie Friedrich is giving a lecture on Ethics and NLP in the Ringvorlesung Alexa, ChatGPT & Co: Wie haltet ihr es mit der Ethik? - Eine interdisziplinäre Perspektive auf KI.
November 13, 2023 -- PhD student Valentin Knappich has participated in the PLABA shared task on simplifying medical text and achieved excellent performance scores both in the automatic and the manual evaluation. Congrats, Valentin! Read more here.
November 8, 2023 -- Alessandra Zarcone is looking for a PhD student for the CHIASM Project at the TH Augsburg, see the call here.
October 27, 2023 -- Alessandra Zarcone's project on "Chancenreiche industrielle Anwendungen für vortrainierte Sprachmodelle (CHIASM)" has been approved for funding, starting January 2024!
October 25, 2023 -- Our paper SciOL and MuLMS-Img: Introducing A Large-Scale Multimodal Scientific Dataset and Models for Image-Text Tasks in the Scientific Domain has been accepted to WACV 2024. The paper describes new datasets and models for language-vision tasks in the scientific domain. This is joint work with the Bosch Center for Artificial Intelligence.
October 10, 2023 -- Our paper Is the Answer in the Text? Challenging ChatGPT with Evidence Retrieval from Instructive Text has been accepted to the Findings of EMNLP 2023. Congrats to the team, Sophie, Talita, and Wei! Our paper MuLMS: A Multi-Layer Annotated Text Corpus for Information Extraction in the Materials Science Domain has been accepted to The 2nd Workshop on Information Extraction from Scientific Publications which will take place at IJCNLP-AACL 2023. Congrats, Timo, Matteo, and Stefan!
October 2, 2023 -- Our paper EduQuick: A Dataset Toward Evaluating Summarization of Informal Educational Content for Social Media has been accepted to Eval4NLP 2023. The paper looks into the use of LLM to summarise educational content and discusses how to evaluate such summaries. This is joint work with the Fraunhofer IIS and OTH Amberg-Weiden.
October 4-5, 2023 -- Our Human Language Technology @ Augsburg Kickoff Workshop took place at Augsburg University.
July 2023 -- Paper on Argumentative Zoning in the materials science domain accepted to CODI@ ACL 2023. Congrats, Timo and Sophie!