Dynamic dialogue modelling

Informazioni relative al progetto

DYMO

ID dell’accordo di sovvenzione: 804636

DOI 10.3030/804636

Data della firma CE 24 Ottobre 2018

Data di avvio 1 Settembre 2019

Data di completamento 31 Gennaio 2026

Finanziato da

• EXCELLENT SCIENCE - European Research Council (ERC)

Costo totale € 1 499 956,00

Contributo UE € 1 499 956,00

Coordinato da

HEINRICH-HEINE-UNIVERSITAET DUESSELDORF
Germany

Pubblicazioni

Conference proceedings (15)

Uncertainty Measures in Neural Belief Tracking and the Effects on Dialogue Policy Performance

Autori: Carel van Niekerk, Andrey Malinin, Christian Geishauser, Michael Heck, Hsien-chin Lin, Nurul Lubis, Shutong Feng and Milica Gašic
Pubblicato in: Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2021, Pagina/e 7901–7914
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2021.emnlp-main.623

LAVA: Latent Action Spaces via Variational Auto-encoding for Dialogue Policy Optimization

Autori: N. Lubis, C. Geishauser, M. Heck, H-C. Lin, M. Moresi, C. van Niekerk, M. Gašić
Pubblicato in: Proceedings of the 28th International Conference on Computational Linguistics (COLING), 2020, Pagina/e 465–479
Editore: International Committee on Computational Linguistics
DOI: 10.18653/v1/2020.coling-main.41

Dialogue Term Extraction using Transfer Learning and Topological Data Analysis

Autori: Renato Vukovic, Michael Heck, Benjamin Ruppik, Carel van Niekerk, Marcus Zibrowius, Milica Gasic
Pubblicato in: 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue, 2022
Editore: Association for Computational Linguistics (ACL)
DOI: 10.5281/zenodo.6858565

Dialogue Evaluation with Offline Reinforcement Learning

Autori: Nurul Lubis, Christian Geishauser, Hsien-Chin Lin, Carel van Niekerk, Michael Heck, Shutong Feng, Milica Gasic
Pubblicato in: 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue, 2022
Editore: Association for Computational Linguistics (ACL)
DOI: 10.48550/arxiv.2209.00876

Robust Dialogue State Tracking with Weak Supervision and Sparse Data

Autori: Michael Heck, Nurul Lubis, Carel van Niekerk, Shutong Feng, Christian Geishauser, Hsien-Chin Lin, Milica Gasic
Pubblicato in: CoRR, 2022
Editore: TBA
DOI: 10.48550/arxiv.2202.03354

Actor-Double-Critic: Incorporating Model-Based Critic for Task-Oriented Dialogue Systems

Autori: Y.-C. Wu, B.-H. Tseng, M. Gašić
Pubblicato in: Findings of EMNLP 2020, 2020
Editore: Association for Computational Linguistics

EmoWOZ: A Large-Scale Corpus and Labelling Scheme for Emotion Recognition in Task-Oriented Dialogue Systems

Autori: Feng, Shutong; Lubis, Nurul; Geishauser, Christian; Lin, Hsien-chin; Heck, Michael; van Niekerk, Carel; Gašić, Milica
Pubblicato in: Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), 2022, Pagina/e 4096–4113
Editore: European Language Resources Association (ELRA)
DOI: 10.48550/arxiv.2109.04919

Knowing What You Know: Calibrating Dialogue Belief State Distributions via Ensembles

Autori: Carel van Niekerk, Michael Heck, Christian Geishauser, Hsien-chin Lin, Nurul Lubis, Marco Moresi, Milica Gasic
Pubblicato in: Findings of the Association for Computational Linguistics: EMNLP 2020, 2020, Pagina/e 3096-3102
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2020.findings-emnlp.277

Out-of-Task Training for Dialog State Tracking Models

Autori: M. Heck, C. Geishauser, H-C. Lin, N. Lubis, M. Moresi, C. van Niekerk, M. Gašić
Pubblicato in: COLING 2020, 2020
Editore: Association for Computational Linguistics

What does the User Want? Information Gain for Hierarchical Dialogue Policy Optimisation

Autori: C. Geishauser, S. Hu, H. Lin, N. Lubis, M. Heck, S. Feng, C. van Niekerk, M. Gasic
Pubblicato in: IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), 2021
Editore: IEEE
DOI: 10.1109/asru51503.2021.9687856

GenTUS: Simulating User Behaviour and Language in Task-oriented Dialogues with Generative Transformers

Autori: Hsien-Chin Lin, Christian Geishauser, Shutong Feng, Nurul Lubis, Carel van Niekerk, Michael Heck and Milica Gasic
Pubblicato in: 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue, 2022
Editore: The Association for Computational Linguistics (ACL)
DOI: 10.48550/arxiv.2208.10817

Dynamic Dialogue Policy for Continual Reinforcement Learning

Autori: Christian Geishauser, Carel van Niekerk, Nurul Lubis, Michael Heck, Hsien-Chin Lin, Shutong Feng, Milica Gasic
Pubblicato in: The 29th International Conference on Computational Linguistics (COLING), 2022
Editore: International Committee on Computational Linguistics
DOI: 10.48550/arxiv.2204.05928

Topology of Word Embeddings: Singularities Reflect Polysemy

Autori: A. Jakubowski, M. Gašić, M. Zibrowius
Pubblicato in: *SEM 2020, 2020
Editore: Association for Computational Linguistics

TripPy: A Triple Copy Strategy for Value Independent Neural Dialog State Tracking

Autori: M. Heck, C. van Niekerk, N. Lubis, C. Geishauser, H-C. Lin, M. Moresi, M. Gašić
Pubblicato in: Proceedings of the 21th Annual Meeting of the Special Interest Group on Discourse and Dialogue (SIGDIAL)), 2020, Pagina/e 35–44
Editore: Association for Computational Linguistics
DOI: 10.48550/arxiv.2005.02877

Domain-independent User Simulation with Transformers for Task-oriented Dialogue Systems

Autori: Lin, Hsien-chin; Lubis, Nurul; Hu, Songbo; van Niekerk, Carel; Geishauser, Christian; Heck, Michael; Feng, Shutong; Gašić, Milica
Pubblicato in: Proceedings of the 22nd Annual Meeting of the Special Interest Group on Discourse and Dialogue (SIGDIAL), 2021, Pagina/e 445–456
Editore: Association for Computational Linguistics
DOI: 10.48550/arxiv.2106.08838