Next-Generation Natural Language Generation

Informazioni relative al progetto

NG-NLG

ID dell’accordo di sovvenzione: 101039303

DOI

10.3030/101039303

Data della firma CE 31 Marzo 2022

Data di avvio 1 Aprile 2022

Data di completamento 31 Marzo 2027

Finanziato da

• European Research Council (ERC)

Costo totale

€ 1 420 375,00

Contributo UE

€ 1 420 375,00

Coordinato da

UNIVERZITA KARLOVA
Czechia

CORDIS fornisce collegamenti ai risultati finali pubblici e alle pubblicazioni dei progetti ORIZZONTE.

I link ai risultati e alle pubblicazioni dei progetti del 7° PQ, così come i link ad alcuni tipi di risultati specifici come dataset e software, sono recuperati dinamicamente da .OpenAIRE .

Pubblicazioni

Conference proceedings (34)

With a Little Help from the Authors: Reproducing Human Evaluation of an MT Error Detector

Autori: Ondřej Plátek, Mateusz Lango, Ondřej Dušek
Pubblicato in: Workshop on Human Evaluation of NLP Systems, Numero 3, 2023, ISBN 978-954-452-088-5
Editore: INCOMA Ltd.
DOI: 10.48550/ARXIV.2308.06527

Two Reproductions of a Human-Assessed Comparative Evaluation of a Semantic Error Detection System

Autori: Huidrom, Rudali; Dušek, Ondřej; Kasner, Zdenek; Castro Ferreira, Thiago; Belz, Anya
Pubblicato in: International Conference on Natural Language Generation: Generation Challenges, Numero 15, 2022, ISBN 978-1-955917-60-5
Editore: Association for Computational Linguistics
DOI: 10.5281/ZENODO.11373776

Tackling Hallucinations in Neural Chart Summarization

Autori: Saad Obaid ul Islam, Iza Škrjanec, Ondřej Dušek, Vera Demberg
Pubblicato in: International Natural Language Generation Conference, Numero 16, 2023, ISBN 979-8-89176-001-1
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2308.00399

AARGH! End-to-end Retrieval-Generation for Task-Oriented Dialog

Autori: Tomáš Nekvinda, Ondřej Dušek
Pubblicato in: Annual Meeting of the Special Interest Group on Discourse and Dialogue, Numero 23, 2022, ISBN 978-1-955917-66-7
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2209.03632

Polite Chatbot: A Text Style Transfer Application

Autori: Sourabrata Mukherjee, Vojtěch Hudeček, Ondřej Dušek
Pubblicato in: Conference of the European Chapter of the Association for Computational Linguistics: Student Research Workshop, Numero 17, 2023, ISBN 978-1-959429-48-7
Editore: Association for Computational Linguistics
DOI: 10.5281/ZENODO.11373712

Frontiers in Artificial Intelligence and Applications

Autori: Jakub Raczyński, Mateusz Lango, Jerzy Stefanowski
Pubblicato in: European Conference on Artificial Intelligence, Numero 26, 2023, ISSN 0922-6389
Editore: IOS Press
DOI: 10.48550/ARXIV.2312.11356

Multilingual Text Style Transfer: Datasets Models for Indian Languages

Autori: Sourabrata Mukherjee, Atul Kr. Ojha, Akanksha Bansal, Deepak Alok, John P. McCrae, Ondřej Dušek
Pubblicato in: International Natural Language Generation Conference, Numero 17, 2024
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2405.20805

Critic-Driven Decoding for Mitigating Hallucinations in Data-to-text Generation

Autori: Mateusz Lango, Ondřej Dušek
Pubblicato in: Conference on Empirical Methods in Natural Language Processing, 2023, ISBN 979-8-89176-060-8
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2310.16964

Semantic Accuracy in Natural Language Generation: A Thesis Proposal

Autori: Patrícia Schmidtová
Pubblicato in: Annual Meeting of the Association for Computational Linguistics, Numero 61, vol. 4, 2023, ISBN 978-1-959429-69-2
Editore: Association for Computational Linguistics
DOI: 10.5281/ZENODO.11373677

MooseNet: A Trainable Metric for Synthesized Speech with a PLDA Module

Autori: Ondřej Plátek; Ondrej Dusek
Pubblicato in: ISCA Speech Synthesis Workshop, Numero 12, 2023
Editore: ISCA
DOI: 10.48550/arxiv.2301.07087

Beyond Traditional Benchmarks: Analyzing Behaviors of Open LLMs on Data-to-Text Generation

Autori: Zdeněk Kasner, Ondřej Dušek
Pubblicato in: Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Numero 62, 2024
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2401.10186

Leak, Cheat, Repeat: Data Contamination and Evaluation Malpractices in Closed-Source LLMs

Autori: Simone Balloccu, Patrícia Schmidtová, Mateusz Lango, and Ondrej Dusek
Pubblicato in: Conference of the European Chapter of the Association for Computational Linguistics, Numero 18, 2024, ISBN 979-8-89176-088-2
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2402.03927

Are LLMs All You Need for Task-Oriented Dialogue?

Autori: Vojtěch Hudeček, Ondřej Dušek
Pubblicato in: Annual Meeting of the Special Interest Group on Discourse and Dialogue, Numero 24, 2023, ISBN 979-8-89176-028-8
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2304.06556

Mind the Labels: Describing Relations in Knowledge Graphs With Pretrained Models

Autori: Zdeněk Kasner, Ioannis Konstas, Ondřej Dušek
Pubblicato in: Conference of the European Chapter of the Association for Computational Linguistics, Numero 17, 2023, ISBN 978-1-959429-44-9
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2210.07373

Are Experts Needed? On Human Evaluation of Counselling Reflection Generation

Autori: Zixiu Wu, Simone Balloccu, Ehud Reiter, Rim Helaoui, Diego Reforgiato Recupero, Daniele Riboni
Pubblicato in: Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2023, ISBN 978-1-959429-72-2
Editore: Association for Computational Linguistics
DOI: 10.18653/V1/2023.ACL-LONG.382

Learning Interpretable Latent Dialogue Actions With Less Supervision

Autori: Vojtěch Hudeček, Ondřej Dušek
Pubblicato in: Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing, Numero 2, 2022, ISBN 978-1-955917-65-0
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2209.11128

WebLINX: Real-World Website Navigation with Multi-Turn Dialogue

Autori: Xing Han Lu, Zdeněk Kasner, Siva Reddy
Pubblicato in: International Conference on Machine Learning, Numero 41, 2024
Editore: PMLR
DOI: 10.48550/ARXIV.2402.05930

LEEETs-Dial: Linguistic Entrainment in End-to-End Task-oriented Dialogue systems

Autori: Nalin Kumar, Ondrej Dusek
Pubblicato in: Findings of the Association for Computational Linguistics: NAACL 2024, 2024
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2311.09390

Low-Resource Text Style Transfer for Bangla: Data & Models

Autori: Sourabrata Mukherjee, Akanksha Bansal, Pritha Majumdar, Atul Kr. Ojha, Ondřej Dušek
Pubblicato in: Workshop on Bangla Language Processing, Numero 1, 2023, ISBN 979-8-89176-058-5
Editore: Association for Computational Linguistics
DOI: 10.5281/ZENODO.11372105

"ReproHum #0043-4: Evaluating Summarization Models: Investigating the Impact of Education and Language Proficiency on Reproducibility"

Autori: Mateusz Lango, Patricia Schmidtová, Simone Balloccu, Ondřej Dušek
Pubblicato in: Workshop on Human Evaluation of NLP Systems, Numero 4, 2024, ISSN 2951-2093
Editore: ELRA and ICCL
DOI: 10.5281/ZENODO.11371369

Ask the experts: sourcing high-quality datasets for nutritional counselling through Human-AI collaboration

Autori: Balloccu, Simone; Reiter, Ehud; Kumar, Vivek; Recupero, Diego Reforgiato; Riboni, Daniele
Pubblicato in: Findings of the Association for Computational Linguistics: EMNLP 2024, 2024
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2401.08420

Leveraging Large Language Models for Building Interpretable Rule-Based Data-to-Text Systems

Autori: Jędrzej Warczyński, Mateusz Lango, Ondřej Dušek
Pubblicato in: International Natural Language Generation Conference, Numero 17, 2024
Editore: Association for Computational Linguistics
DOI: 10.5281/ZENODO.14236701

VisuaLLM: Easy Web-based Visualization for Neural Language Generation

Autori: František Trebuňa, Ondřej Dušek
Pubblicato in: International Natural Language Generation Conference: System Demonstrations, Numero 16, 2023, ISBN 979-8-89176-002-8
Editore: Association for Computational Linguistics
DOI: 10.5281/ZENODO.11372124

Generating clickbait spoilers with an ensemble of large language models

Autori: Mateusz Woźny, Mateusz Lango
Pubblicato in: International Natural Language Generation Conference, Numero 16, 2023, ISBN 979-8-89176-001-1
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2404.15758

Neural Pipeline for Zero-Shot Data-to-Text Generation

Autori: Zdeněk Kasner, Ondřej Dušek
Pubblicato in: Annual Meeting of the Association for Computational Linguistics, Numero 60, 2022, ISBN 978-1-955917-21-6
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2203.16279

Are Large Language Models Actually Good at Text Style Transfer?

Autori: Sourabrata Mukherjee, Atul Kr. Ojha, Ondřej Dušek
Pubblicato in: International Natural Language Generation Conference, Numero 17, 2024
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2406.05885

Balancing the Style-Content Trade-Off in Sentiment Transfer Using Polarity-Aware Denoising

Autori: Sourabrata Mukherjee, Zdeněk Kasner, Ondřej Dušek
Pubblicato in: Text, Speech, and Dialogue, 2022, ISBN 9783031162695
Editore: Springer International Publishing
DOI: 10.48550/ARXIV.2312.14708

Leveraging Low-resource Parallel Data for Text Style Transfer

Autori: Sourabrata Mukherjee, Ondřej Dušek
Pubblicato in: International Natural Language Generation Conference, Numero 16, 2023, ISBN 979-8-89176-001-1
Editore: Association for Computational Linguistics
DOI: 10.5281/ZENODO.11373566

Automatic Metrics in Natural Language Generation: A survey of Current Evaluation Practices

Autori: Patrícia Schmidtová, Saad Mahamood, Simone Balloccu, Ondřej Dušek, Albert Gatt, Dimitra Gkatzia, David M. Howcroft, Ondřej Plátek, Adarsa Sivaprasad
Pubblicato in: International Natural Language Generation Conference, Numero 17, 2024
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2408.09169

Three Ways of Using Large Language Models to Evaluate Chat

Autori: Ondřej Plátek, Vojtěch Hudeček, Patricia Schmidtová, Mateusz Lango, Ondřej Dušek
Pubblicato in: Dialog System Technology Challenge, Numero 11, 2023
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2308.06502

Faithful and Plausible Natural Language Explanations for Image Classification: A Pipeline Approach

Autori: Adam Wojciechowski, Mateusz Lango, Ondrej Dušek
Pubblicato in: Findings of the Association for Computational Linguistics: EMNLP 2024, 2024
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2407.20899

TabGenie: A Toolkit for Table-to-Text Generation

Autori: Zdeněk Kasner, Ekaterina Garanina, Ondřej Plátek, Ondřej Dušek
Pubblicato in: Annual Meeting of the Association for Computational Linguistics, Numero 61, 2023
Editore: 978-1-959429-70-8
DOI: 10.48550/ARXIV.2302.14169

Better Translation + Split and Generate for Multilingual RDF-to-Text (WebNLG 2023)

Autori: Nalin Kumar, Saad Obaid Ul Islam, Ondřej Dušek
Pubblicato in: Workshop on Multimodal, Multilingual Natural Language Generation and Multilingual WebNLG Challenge, 2023
Editore: Association for Computational Linguistics
DOI: 10.5281/ZENODO.11373606

factgenie: A Framework for Span-based Evaluation of Generated Texts

Autori: Zdeněk Kasner, Ondřej Plátek, Patrícia Schmidtová, Simone Balloccu, Ondřej Dušek
Pubblicato in: International Natural Language Generation Conference: System Demonstrations, Numero 17, 2024
Editore: Association for Computational Linguistics
DOI: 10.48550/ARXIV.2407.17863

Peer reviewed articles (1)

Barriers and enabling factors for error analysis in NLG research

Autori: Emiel Van Miltenburg; Miruna Clinciu; Ondřej Dušek; Dimitra Gkatzia; Stephanie Inglis; Leo Leppänen; Saad Mahamood; Stephanie Schoch; Craig Thomson; Luou Wen
Pubblicato in: Northern European Journal of Language Technology, 2023, ISSN 2000-1533
Editore: Northern European Association of Language Technology
DOI: 10.3384/nejlt.2000-1533.2023.4529