Zum Inhalt springen Zur Suche springen

Forschung des Lehrstuhls Dialog Systems and Machine Learning

2026

  • Text-to-SQL Task-oriented Dialogue Ontology Construction, Renato Sprenger-Vukovic, ACL Conference, San Diego, USA, 2026
  • (Very personal) view of the development of AI and dialogue systems, Milica Gasic, Wikipedia conference, Belgrade, SRB, 2026
  • Resources for dialogue systems - from corpora to LLMs, Milica Gasic, University of Belgrade, SRB, 2026
  • Local Intrinsic Dimension Estimates of Word Embedding Spaces, Benjamin Ruppik, University of Fribourg, CH, 2026
  • Analogies, pain and geometry of intelligence - three views on (unsupervised) learning, Milica Gasic, Serbian Academy of Sciences and Arts, Belgrade, SRB, 2026
  • Measuring the Topology of Word Embeddings, Benjamin Ruppik, TDA Seminar of the Dioscuri Centre in Topological Data Analysis, Warsaw and virtual, PL, 2026
  • Competence, Pain and the Geometry of Intelligence: Three Perspectives on Learning and Representation, Milica Gasic, Natural Language & Information Processing (NLIP) seminar at the University of Cambridge, UK, 2026

2025

  • Less is More: Local Intrinsic Dimensions of Contextual Language Models, Benjamin Ruppik, NeurIPS Conference, San Diego, USA, 2025
  • Measuring the Topology of Word Embeddings, Benjamin Ruppik, KTH, Stockholm, SE, 2025
  • A Confidence-based Acquisition Model for Selfsupervised Active Learning and Label Correction, Carel van Niekerk, ACL conference, Vienna, AT, 2025
  • Keynote Lecture: Natural Language Processing, Milica Gasic, University of Sarajevo, Sarajevo, BA, 2025
  • ConvLab-3: A Flexible Dialogue System Toolkit Based on a Unified Data Format, Nurul Lubis, SIGIR Workshop on Simulations for Information Access (Sim4IA), Padua, IT, 2025
  • Topological Deep Learning for Term Extraction, Benjamin Ruppik, HHU, Düsseldorf, DE, 2025
  • Learning With an Open Horizon in Ever-Changing Dialogue Circumstances, Hsien-Chin Lin, IEEE International Conference on Acoustics, Speech, and Signal Processing, Hyderabad, IN, 2025
  • Learning from Noisy Labels via Self-Taught On-the-Fly Meta Loss Rescaling, Michael Heck, The 39th AAAI Conference on Artificial Intelligence, Philadelphia, USA, 2025
  • Text as a parameter, Hsien-Chin Lin, TU Dortmund, Dortmund, DE, 2025
  • Sources of understanding and confusion in conversational systems and how to learn from them, Milica Gasic, Fondazione Bruno Kessler, Trento, IT, 2025

2024

  • Infusing Emotions into Task-oriented Dialogue Systems: Understanding, Management, and Generation, Shutong Feng, SIGDIAL Conference, Kyoto, JP, 2024
  • Dialogue Ontology Relation Extraction using Constrained Chain-of-Thought Decoding, Renato Vukovic, SIGDIAL Conference and 20th YRRSDS workshop, Kyoto, JP, 2024
  • Topological Deep Learning for Term Extraction, Benjamin Ruppik, 20th YRRSDS workshop, Kyoto, JP, 2024
  • Local Topology Measures of Contextual Language Model Latent Spaces With Applications to Dialogue Term Extraction; Benjamin Ruppik, SIGDIAL Conference, Kyoto, JP, 2024
  • Leaning with an open horizon, Milica Gasic, Apple, Cupertino, USA, 2024
  • Is a large language model a (new) model for dialogue?, Milica Gasic, Trivago AI Meet-up, Düsseldorf, DE, 2024
  • Generative Artificial Intelligence, Hsien-Chin Lin, HHU HeiCAD Retreat, Düsseldorf, DE
  • Why LLMs need to worry?, Milica Gasic, Bamberger Zentrum für Künstliche Intelligenz, BaCAI Lectures, Bamberg, DE, 2024
  • Slow and steady wins the race: Reinforcement Learning in Conversational AI, Milica Gasic, Pint of Science Festival, Düsseldorf, DE, 2024
  • Continually learning Conversational AI, Milica Gasic, The 5th International Symposium on AI and Electronics, Tohoku University, Tohoku, JP, 2024
  • Continually learning Conversational AI, Milica Gasic, LT-Bridge Winter School, DKFI Saarbrücken, DE, 2024
  • Understanding Conversational Al, Milica Gasic, HHU Bürgeruni "Forschung im Fokus", Düsseldorf, DE, 2024

2023

  • ConvLab-3: A Flexible Dialogue System Toolkit Based on a Unified Data Format, Christian Geishauser and Carel van Niekerk, EMNLP, Singapore, SG, 2023
  • From Chatter to Matter: Addressing Critical Steps of Emotion Recognition Learning in Task-oriented Dialogue, Shutong Feng, SIGDIAL conference, Prague, CZ, 2023
  • Towards Emotion-aware Task-oriented Dialogue Systems, Shutong Feng, Cambridge, UK, and Helsinki, FI, 2023
  • Beyond Hype: Technical Insights into ChatGPT and the Future of Conversational AI, Renato Vukovic, HHU HeiCAD Brown Bag Lunch, Düsseldorf, DE, 2023, Aligned_LLMs_HeiCAD_presentation
  • Explaining the Sensation: An Accessible Introduction to ChatGPT, Michael Heck, HHU HeiCAD Special Event, Düsseldorf, DE, 2023 ChatGPT2023
  • Dialogue Evaluation via Offline Reinforcement Learning and Emotion Prediction, Nurul Lubis, Inivited talk at the 5th Workshop on NLP for Conversational AI, Toronto, CA, 2023 Evaluation-072023_convai5
  • Robust Dialogue State Tracking with Weak Supervision and Sparse Data, Michael Heck, The 61st Annual Meeting of the Association for Computational Linguistics (ACL), Toronto, CA, 2023
  • Dialogue Evaluation via Offline Reinforcement Learning and Emotion Prediction, Nurul Lubis, Keynote at the 32. Tagung der Computerlinguistik-Studierenden (TaCoS), Düsseldorf, DE, 2023 Evaluation-062023_tacos
  • EmoUS: Simulating User Emotions in Task-Oriented Dialogues, Hsien-Chin Lin, 46th International ACM SIGIR Conference, Taipei, TW, 2023
  • Dynamic Dialogue Policy for Continual Reinforcement Learning, Christian Geishauser, Reinforcement Learning Summer School (RLSS), Barcelona, ES, 2023
  • Exploring the Shape of Word Spaces with Topological Data Analysis, Benjamin Ruppik, 1) Pitt NLP Seminar, University of Pittsburgh Computer Science department, Pittsburgh, USA 2) MIT CSAIL Spoken Language Systems Group, Cambridge, USA 3) NLP Seminar, Columbia University, New York, USA, 2023
  • User simulation in task-oriented dialogue systems, Hsien-Chin Lin, National Taiwan University, Taipei, TW, 2023
  • Aligned Large Language Models, Renato Vukovic, HHU - Heine Center for Artificial Intelligence and Data Science (HeiCAD), Düsseldorf, DE, 2023

2022

  • Topological Data Analysis in Word Embedding Spaces, Benjamin Ruppik, Invited talk at the Geometry graduate colloquium, ETH Zurich, CH, 2022
  • Dynamic Dialogue Policy for Continual Reinforcement Learning, Christian Geishauser, 29th International Conference on Computational Linguistics (COLING), Gyeongju, KR, 2022
  • Detecting Relevant Terms in Word Embedding Spaces, Benjamin Ruppik, 3rd Workshop on Topological Methods in Data Analysis Heidelberg University, Heidelberg, DE, 2022
  • Dialogue Evaluation with Offline Reinforcement Learning, Nurul Lubis, SIGDIAL conference, Edinburgh and virtual, UK, 2022
  • Topology of Word Embedding Spaces, Benjamin Ruppik, 18th YRRSDS workshop, Edinburgh, UK, 2022
  • GenTUS: Simulating User Behaviour and Language in Task-oriented Dialogues with Generative Transformers, Hsien-Chin Lin, SIGDIAL conference and YRRSDS workshop, Edinburgh, UK, 2022
  • Dialogue Term Extraction using Transfer Learning and Topological Data Analysis, Renato Vukovic and Benjamin Ruppik, SIGDIAL conference, Edinburgh, UK, 2022
  • Dialogue Term Extraction using Transfer Learning and Topological Data Analysis, Renato Vukovic, 18th YRRSDS workshop, Edinburgh, UK, 2022
  • EmoWOZ: A Large-Scale Corpus and Labelling Scheme for Emotion Recognition in Task-Oriented Dialogue Systems, Shutong Feng, LREC conference, Marseille, FR, 2022

2021

  • Towards Adaptive and Evolving Task-oriented Dialogue Systems, Michael Heck, Fraunhofer IAIS, virtual, 2021
  • What Does The User Want? Information Gain for Hierarchical Dialogue Policy Optimisation, Christian Geishauser, IEEE ASRU, Cartagena and virtual, CO, 2021
  • Uncertainty Measures in Neural Belief Tracking and the Effects on Dialogue Policy Performance, Carel van Niekerk, EMNLP conference, virtual, 2021
  • Shifts and other vices of human-computer dialogues, Nurul Lubis and Michael Heck, Shifts Challenge@NeurIPS, virtual, 2021
  • Towards Adaptive and Evolving Task-oriented Dialogue Systems, Nurul Lubis and Michael Heck, NAIST SLP Symposium, Nara and virtual, JP, 2021
  • Neural Dialogue Models: Growing up confident and independent, Milica Gasic, KONVENS conference, Düsseldorf, DE, 2021
  • User-centric dialogue models, Milica Gasic, NSF Future Directions Workshop, Emory NLP Research Lab, virtual, 2021
  • Towards Dynamic Dialogue Models, Milica Gasic, Annual Conference of the CDT in Speech and Language Technologies (SLTs), University of Sheffield, virtual, 2021
  • A topological view of polysemy, Milica Gasic, Lectures on Computational Linguistics, Italian Association for Computational Linguistics, University of Roma, Rome and virtual, IT, 2021
  • Towards Dynamic Dialogue Models, Milica Gasic, TU Darmstadt, virtual, 2021
  • What Does the User Want? Information Gain for Hierarchical Dialogue Policy Optimisation, Christian Geishauser, YRRSDS workshop, Singapore and virtual, SG, 2021
  • Knowing What You Know: Calibrating Dialogue Belief State Distributions via Ensembles, Carel van Niekerk, YRRSDS workshop, Singapore and virtual, SG, 2021
  • Domain-independent User Simulation with Transformers for Task-oriented Dialogue Systems, Hsien-Chin Lin, SIGDIAL conference and YRRSDS workshop, Singapore and virtual, SG, 2021

2020

  • Dialogue as a sequential decision-making process, Milica Gasic, ML in PL Conference, virtual, 2020
  • Out-of-Task Training for Dialog State Tracking Models, Michael Heck, COLING conference, virtual, 2020
  • LAVA: Latent Action Spaces via Variational Autoencoding for Dialogue Policy Optimization, Nurul Lubis, COLING conference, virtual, 2020
  • Dialogue as a sequential decision-making process, Milica Gasic, HHU HeiCAD AI Conference, virtual, 2020
  • 10 things you should know about dialogue, Milica Gasic, 7th Frederick Jelinek Memorial Summer Workshop on Speech and Language Technology, JSALT, virtual, 2020
  • TripPy: A Triple Copy Strategy for Value Independent Neural Dialog State Tracking, Michael Heck, SIGDIAL conference, virtual, 2020
  • On the track of multi-domain dialogue models, Milica Gasic, NeurIPS Human in the loop workshop, virtual, 2020

2019

2024

  • Slow and steady wins the race: Reinforcement learning in Conversational AI, Pint of Science, Düsseldorf, 2024

2023

2022