Incorporating Demographic Factors into Natural Language Processing Models

Informations projet

INTEGRATOR

N° de convention de subvention: 949944

DOI

10.3030/949944

Date de signature de la CE 13 Novembre 2020

Date de début 1 Mars 2021

Date de fin 31 Août 2026

Financé au titre de

• EXCELLENT SCIENCE - European Research Council (ERC)

Coût total

€ 1 498 937,00

Contribution de l’UE

€ 1 498 937,00

Coordonné par

UNIVERSITA COMMERCIALE LUIGI BOCCONI
Italy

CORDIS fournit des liens vers les livrables publics et les publications des projets HORIZON.

Les liens vers les livrables et les publications des projets du 7e PC, ainsi que les liens vers certains types de résultats spécifiques tels que les jeux de données et les logiciels, sont récupérés dynamiquement sur OpenAIRE .

Livrables

Open Research Data Pilot (1)

Data Management Plan (DMP)

Data Management Plan of the project

Publications

Conference proceedings (28)

SocioProbe: What, When, and Where Language Models Learn about Sociodemographics

Auteurs: Lauscher, Anne; Bianchi, Federico; Bowman, Samuel R.; Hovy, Dirk
Publié dans: Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, 2022, Page(s) 7901–7918
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2022.emnlp-main.539

XLM-EMO: Multilingual Emotion Prediction in Social Media Text

Auteurs: Federico Bianchi, Debora Nozza, Dirk Hovy
Publié dans: Proceedings of the 12th Workshop on Computational Approaches to Subjectivity, Sentiment & Social Media Analysis, 2022, Page(s) 195–203
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2022.wassa-1.18

Hard and Soft Evaluation of NLP models with BOOtSTrap SAmpling - BooStSa

Auteurs: Tommaso Fornaciari; Alexandra Uma; Massimo Poesio; Dirk Hovy
Publié dans: Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics: System Demonstrations, 2022, Page(s) 127–134
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2022.acl-demo.12

Fair and Argumentative Language Modeling for Computational Argumentation

Auteurs: Carolin Holtermann, Anne Lauscher, Simone Ponzetto
Publié dans: Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Numéro 1, 2022, Page(s) 7841–7861
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2022.acl-long.541

DS-TOD: Efficient Domain Specialization for Task-Oriented Dialog

Auteurs: Chia-Chien Hung; Anne Lauscher; Simone Ponzetto; Goran Glavaš
Publié dans: Findings of the Association for Computational Linguistics: ACL 2022, Numéro 1, 2022, Page(s) 891–904
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2022.findings-acl.72

Pipelines for Social Bias Testing of Large Language Models

Auteurs: Debora Nozza; Federico Bianchi; Dirk Hovy
Publié dans: "Proceedings of BigScience Episode #5 -- Workshop on Challenges & Perspectives in Creating Large Language Models", Numéro 1, 2022, Page(s) 68–74
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2022.bigscience-1.6

Measuring Harmful Sentence Completion in Language Models for LGBTQIA+ Individuals

Auteurs: Debora Nozza; Federico Bianchi; Anne Lauscher; Dirk Hovy
Publié dans: Proceedings of the Second Workshop on Language Technology for Equality, Diversity and Inclusion, 2022, Page(s) 26–34
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2022.ltedi-1.4

SafetyKit: First Aid for Measuring Safety in Open-domain Conversational Systems

Auteurs: Emily Dinan; Gavin Abercrombie; A. Bergman; Shannon Spruit; Dirk Hovy; Y-Lan Boureau; Verena Rieser
Publié dans: Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2022, Page(s) 4113–4133
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2022.acl-long.284

Two Contrasting Data Annotation Paradigms for Subjective NLP Tasks

Auteurs: Paul Rottger; Bertie Vidgen; Dirk Hovy; Janet Pierrehumbert
Publié dans: Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 2022, Page(s) 175–190
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2022.naacl-main.13

Computer says “No”: The Case Against Empathetic Conversational AI

Auteurs: Alba Cercas Curry, Amanda Cercas Curry
Publié dans: Findings of the Association for Computational Linguistics: ACL 2023, 2023, Page(s) 8123–8130
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2023.findings-acl.515

MilaNLP at SemEval-2023 Task 10: Ensembling Domain-Adapted and Regularized Pretrained Language Models for Robust Sexism Detection

Auteurs: Amanda Cercas Curry, Giuseppe Attanasio, Debora Nozza, Dirk Hovy
Publié dans: Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023), 2023, Page(s) 2067–2074
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2023.semeval-1.285

What about “em”? How Commercial Machine Translation Fails to Handle (Neo-)Pronouns

Auteurs: Anne Lauscher, Debora Nozza, Ehm Miltersen, Archie Crowley, Dirk Hovy
Publié dans: Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Numéro 1, 2023, Page(s) 377–392
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2023.acl-long.23

Exposing the limits of Zero-shot Cross-lingual Hate Speech Detection

Auteurs: Debora Nozza
Publié dans: Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers), Numéro 2, 2021, Page(s) 907–914
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2021.acl-short.114

Bridging Fairness and Environmental Sustainability in Natural Language Processing

Auteurs: Marius Hessenthaler, Emma Strubell, Dirk Hovy, Anne Lauscher
Publié dans: Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, 2022, Page(s) 7817–7836
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2022.emnlp-main.533

Temporal and Second Language Influence on Intra-Annotator Agreement and Stability in Hate Speech Labelling

Auteurs: Gavin Abercrombie, Dirk Hovy, Vinodkumar Prabhakaran
Publié dans: Proceedings of the 17th Linguistic Annotation Workshop (LAW-XVII), 2023, Page(s) 96–103
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2023.law-1.10

Entropy-based Attention Regularization Frees Unintended Bias Mitigation from Lists

Auteurs: Giuseppe Attanasio; Debora Nozza; Dirk Hovy; Elena Baralis
Publié dans: Findings of the Association for Computational Linguistics: ACL 2022, 2022, Page(s) 1105–1119
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2022.findings-acl.88

Can Demographic Factors Improve Text Classification? Revisiting Demographic Adaptation in the Age of Transformers

Auteurs: Chia-Chien Hung, Anne Lauscher, Dirk Hovy, Simone Paolo Ponzetto, Goran Glavaš
Publié dans: Findings of the Association for Computational Linguistics: EACL 2023, 2023, Page(s) 1565–1580
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2023.findings-eacl.116

HONEST: Measuring Hurtful Sentence Completion in Language Models

Auteurs: Debora Nozza; Federico Bianchi; Dirk Hovy
Publié dans: Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Numéro 1, 2021, Page(s) 2398–2406
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2021.naacl-main.191

Data-Efficient Strategies for Expanding Hate Speech Detection into Under-Resourced Languages

Auteurs: Röttger, Paul; Nozza, Debora; Bianchi, Federico; Hovy, Dirk
Publié dans: Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, 2022, Page(s) 5674–5691
Éditeur: Association for Computational Linguistics Abu Dhabi, United Arab Emirates
DOI: 10.18653/v1/2022.emnlp-main.383

The Importance of Modeling Social Factors of Language: Theory and Practice

Auteurs: Dirk Hovy; Diyi Yang
Publié dans: Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Numéro 1, 2021, Page(s) 588–602
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2021.naacl-main.49

Language Invariant Properties in Natural Language Processing

Auteurs: Federico Bianchi; Debora Nozza; Dirk Hovy
Publié dans: Proceedings of NLP Power! The First Workshop on Efficient Benchmarking in NLP, 2022, Page(s) 84–92
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2022.nlppower-1.9

Benchmarking Post-Hoc Interpretability Approaches for Transformer-based Misogyny Detection

Auteurs: Giuseppe Attanasio; Debora Nozza; Eliana Pastor; Dirk Hovy
Publié dans: Proceedings of NLP Power! The First Workshop on Efficient Benchmarking in NLP, 2022, Page(s) 100–112
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2022.nlppower-1.11

The State of Profanity Obfuscation in Natural Language Processing Scientific Publications

Auteurs: Debora Nozza, Dirk Hovy
Publié dans: Findings of the Association for Computational Linguistics: ACL 2023, 2023, Page(s) 3897–3909
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2023.findings-acl.240

Twitter-Demographer: A Flow-based Tool to Enrich Twitter Data

Auteurs: Federico Bianchi, Vincenzo Cutrona, Dirk Hovy
Publié dans: Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, 2022, Page(s) 289–297
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2022.emnlp-demos.29

Sustainable Modular Debiasing of Language Models

Auteurs: Anne Lauscher; Tobias Lueken; Goran Glavaš
Publié dans: Findings of the Association for Computational Linguistics: EMNLP 2021, 2021, Page(s) 4782–4797
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2021.findings-emnlp.411

The Ecological Fallacy in Annotation: Modeling Human Label Variation goes beyond Sociodemographics

Auteurs: Matthias Orlikowski, Paul Röttger, Philipp Cimiano, Dirk Hovy
Publié dans: Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), Numéro 2, 2023, Page(s) 1017–1029
Éditeur: Association for Computational Linguistics
DOI: 10.18653/v1/2023.acl-short.88

Welcome to the modern world of pronouns: identity-inclusive Natural Language Processing beyond gender

Auteurs: Lauscher, Anne; Crowley, Archie; Hovy, Dirk
Publié dans: Proceedings of the 29th International Conference on Computational Linguistics, 2022, Page(s) 1221–1232
Éditeur: International Committee on Computational Linguistics

Guiding the Release of Safer E2E Conversational AI through Value Sensitive Design

Auteurs: Bergman, A. Stevie; Abercrombie, Gavin; Spruit, Shannon; Hovy, Dirk; Dinan, Emily; Boureau, Y-Lan; Verena Rieser
Publié dans: Proceedings of the 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue, 2022, Page(s) 39–52
Éditeur: Association for Computational Linguistics

Peer reviewed articles (2)

Five sources of bias in natural language processing

Auteurs: Dirk Hovy, Shrimai Prabhumoye
Publié dans: Language and Linguistics Compass, Wiley, 2021, ISSN 1749-818X
Éditeur: Blackwell
DOI: 10.1111/lnc3.12432

Viewpoint: Artificial Intelligence Accidents Waiting to Happen?

Auteurs: Federico Bianchi, Amanda Cercas Curry, Dirk Hovy
Publié dans: Journal of Artificial Intelligence Research, Numéro Vol. 76 (2023), 2023, ISSN 1076-9757
Éditeur: Morgan Kaufmann Publishers, Inc.
DOI: 10.1613/jair.1.14263