INTEGRATOR
INTEGRATOR logo
Dirk Hovy, scientific director of DMI and Professor of computer science, has won an ERC starting grant of 1.5mln euros. His project INTEGRATOR, funded under grant agreement 949944, introduces demographic factors into language processing systems, which will improve algorithmic performance, avoid racism, sexism, and ageism, and open up new applications. What if I wrote that “winning an ERC Grant, Dirk Hovy got a sick result?”. Those familiar with the use of “sick” as a synonym for “great” or “awesome” among teenagers would think that Bocconi Knowledge hired a very young writer (or someone posing as such). The rest would think I went crazy. Current artificial intelligence-based language systems wouldn’t have a clue. “Natural language processing (NLP) technologies,” Prof. Hovy says, “fail to account for demographics both in understanding language and in generating it. And this failure prevents us from reaching human-like performance. It limits possible future applications and it introduces systematic bias against underrepresented demographic groups”.
🗞️🗞️ Related articles featured in Corriere Innovazione and Bocconi News.
Publications
Consistency is Key: Disentangling Label Variation in Natural Language Processing with Intra-Annotator Agreement
Proceedings of the The 4th Workshop on Perspectivist Approaches to NLP
November, 2025
November, 2025
We commonly use agreement measures to assess the utility of judgements made by human annotators in Natural Language Processing (NLP) …
No for Some, Yes for Others: Persona Prompts and Other Sources of False Refusal in Language Models
Proceedings of the 9th Widening NLP Workshop
November, 2025
November, 2025
Large language models (LLMs) are increasingly integrated into our daily lives and personalized. However, LLM personalization might also …
IssueBench: Millions of Realistic Prompts for Measuring Issue Bias in LLM Writing Assistance
TACL
September, 2025
September, 2025
Large language models (LLMs) are helping millions of users write texts about diverse issues, and in doing so expose users to different …
The AI Gap: How Socioeconomic Status Affects Language Technology Interactions
ACL 2025
July, 2025
July, 2025
Socioeconomic status (SES) fundamentally influences how people interact with each other and, more recently, with digital technologies …
Beyond Demographics: Fine-tuning Large Language Models to Predict Individuals' Subjective Text Perceptions
Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
July, 2025
July, 2025
People naturally vary in their annotations for subjective questions and some of this variation is thought to be due to the …
Educators' Perceptions of Large Language Models as Tutors: Comparing Human and AI Tutors in a Blind Text-only Setting
Proceedings of the 20th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2025)
July, 2025
July, 2025
The rapid development of Large Language Models (LLMs) opens up the possibility of using them aspersonal tutors. This has led to the …
Twists, Humps, and Pebbles: Multilingual Speech Recognition Models Exhibit Gender Performance Gaps
Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
November, 2024
November, 2024
Current automatic speech recognition (ASR) models are designed to be used across many languages and tasks without substantial changes. …
Countering Hateful and Offensive Speech Online - Open Challenges
EMNLP 2024 Tutorials Abstracts
September, 2024
September, 2024
In today’s digital age, hate speech and offensive speech online pose a significant challenge to maintaining respectful and inclusive …
Divine LLaMAs: Bias, Stereotypes, Stigmatization, and Emotion Representation of Religion in Large Language Models
EMNLP 2024
September, 2024
September, 2024
Emotions play important epistemological and cognitive roles in our lives, revealing our values and guiding our actions. Previous work …
Compromesso! Italian Many-Shot Jailbreaks Undermine the Safety of Large Language Models
ACL 2024 (SRW)
August, 2024
August, 2024
As diverse linguistic communities and users adopt large language models (LLMs), assessing their safety across languages becomes …
My Answer is C: First-Token Probabilities Do Not Match Text Answers in Instruction-Tuned Language Models
ACL 2024 (Findings)
August, 2024
August, 2024
The open-ended nature of language generation makes the evaluation of autoregressive large language models (LLMs) challenging. One …
XSTest: A Test Suite for Identifying Exaggerated Safety Behaviors in Large Language Models
NAACL 2024 (Main)
July, 2024
July, 2024
Without proper safeguards, large language models will readily follow malicious instructions and generate toxic content. This risk …
Beyond Flesch-Kincaid: Prompt-based Metrics Improve Difficulty Classification of Educational Texts
arXiv
May, 2024
May, 2024
Using large language models (LLMs) for educational applications like dialogue-based teaching is a hot topic. Effective teaching, …
Wisdom of Instruction-Tuned Language Model Crowds. Exploring Model Label Variation
Proceedings of the 3rd Workshop on Perspectivist Approaches to NLP (NLPerspectives) @ LREC-COLING 2024
May, 2024
May, 2024
Large Language Models (LLMs) exhibit remarkable text classification capabilities, excelling in zero- and few-shot learning (ZSL and …
Safety-Tuned LLaMAs: Lessons From Improving the Safety of Large Language Models that Follow Instructions
ICLR 2024
May, 2024
May, 2024
Training large language models to follow instructions makes them perform better on a wide range of tasks, generally becoming more …
SafetyPrompts: a Systematic Review of Open Datasets for Evaluating and Improving Large Language Model Safety
arxiv
April, 2024
April, 2024
The last two years have seen a rapid growth in concerns around the safety of large language models (LLMs). Researchers and …
Angry Men, Sad Women: Large Language Models Reflect Gendered Stereotypes in Emotion Attribution
ACL 2024
March, 2024
March, 2024
Large language models (LLMs) reflect societal norms and biases, especially about gender. While societal biases and stereotypes have …
Conversations as a Source for Teaching Scientific Concepts at Different Education Levels
arXiv
March, 2024
March, 2024
Open conversations are one of the most engaging forms of teaching. However, creating those conversations in educational software is a …
Emotion Analysis in NLP: Trends, Gaps and Roadmap for Future Directions
LREC-COLING 2024
March, 2024
March, 2024
Emotions are a central aspect of communication. Consequently, emotion analysis (EA) is a rapidly growing field in natural language …
Explaining Speech Classification Models via Word-Level Audio Segments and Paralinguistic Features
Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)
March, 2024
March, 2024
Predictive models make mistakes and have biases. To combat both, we need to understand their predictions. Explainable AI (XAI) …
Classist Tools: Social Class Correlates with Performance in NLP
arXiv
March, 2024
March, 2024
Since the foundational work of William Labov on the social stratification of language (Labov, 1964), linguistics has made concentrated …
Impoverished Language Technology: The Lack of (Social) Class in NLP
LREC-COLING
March, 2024
March, 2024
Since Labov’s (1964) foundational work on the social stratification of language, linguistics has dedicated concerted efforts …
Subjective isms? On the Danger of Conflating Hate and Offence in Abusive Language Detection
WOAH24
March, 2024
March, 2024
Natural language processing research has begun to embrace the notion of annotator subjectivity, motivated by variations in labelling. …
A Tale of Pronouns: Interpretability Informs Gender Bias Mitigation for Fairer Instruction-Tuned Machine Translation
EMNLP 2023
December, 2023
December, 2023
Recent instruction fine-tuned models can solve multiple NLP tasks when prompted to do so, with machine translation (MT) being a …
Mirages. On Anthropomorphism in Dialogue Systems
EMNLP23
December, 2023
December, 2023
Automated dialogue or conversational systems are anthropomorphised by developers and personified by users. While a degree of …
The Empty Signifier Problem: Towards Clearer Paradigms for Operationalising 'Alignment' in Large Language Models
SoLaR at NeurIPs 2023
November, 2023
November, 2023
In this paper, we address the concept of ‘alignment’ in large language models (LLMs) through the lens of post-structuralist …
SimpleSafetyTests: a Test Suite for Identifying Critical Safety Risks in Large Language Models
arXiv
November, 2023
November, 2023
The past year has seen rapid acceleration in the development of large language models (LLMs). For many tasks, there is now a wide range …
The Past, Present and Better Future of Feedback Learning in Large Language Models for Subjective Human Preferences and Values
EMNLP 2023
October, 2023
October, 2023
Human feedback is increasingly used to steer the behaviours of Large Language Models (LLMs). However, it is unclear how to collect and …
Wisdom of Instruction-Tuned Language Model Crowds: Exploring Model Label Variation
NLPerspectives (LREC-COLING 2024)
July, 2023
July, 2023
Large Language Models (LLMs) exhibit remarkable text classification capabilities, excelling in zero- and few-shot learning (ZSL and …
MilaNLP at SemEval-2023 Task 10: Ensembling Domain-Adapted and Regularized Pretrained Language Models for Robust Sexism Detection
SemEval-2023
July, 2023
July, 2023
We present the system proposed by the MilaNLP team for the Explainable Detection of Online Sexism (EDOS) shared task. We propose an …
Respectful or Toxic? Using Zero-Shot Learning with Language Models to Detect Hate Speech
The 7th Workshop on Online Abuse and Harms (WOAH) at ACL 2023
July, 2023
July, 2023
Hate speech detection faces two significant challenges: 1) the limited availability of labeled data and 2) the high variability of hate …
Temporal and Second Language Influence on Intra-Annotator Agreement and Stability in Hate Speech Labelling
Proceedings of LAW-XVII
July, 2023
July, 2023
Much work in natural language processing (NLP) relies on human annotation. The majority of this implicitly assumes that annotator’s …
The Ecological Fallacy in Annotation: Modeling Human Label Variation goes beyond Sociodemographics
Findings of ACL 2023
July, 2023
July, 2023
Many NLP tasks exhibit human label variation, where different annotators give different labels to the same texts. This variation is …
The State of Profanity Obfuscation in Natural Language Processing Scientific Publications
Findings of the Association for Computational Linguistics: ACL 2023
July, 2023
July, 2023
Work on hate speech has made considering rude and harmful examples in scientific publications inevitable. This situation raises various …
What about ''em''? How Commercial Machine Translation Fails to Handle (Neo-)Pronouns
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
July, 2023
July, 2023
As 3rd-person pronoun usage shifts to include novel forms, e.g., neopronouns, we need more research on identity-inclusive NLP. …
What about ''em''? How Commercial Machine Translation Fails to Handle (Neo-)Pronouns
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
July, 2023
July, 2023
As 3rd-person pronoun usage shifts to include novel forms, e.g., neopronouns, we need more research on identity-inclusive NLP. …
Computer says “No”: The Case Against Empathetic Conversational AI
ACL Findings 2023
June, 2023
June, 2023
Emotions are an integral part of human cognition and they guide not only our understanding of the world but also our actions within it. …
Can Demographic Factors Improve Text Classification? Revisiting Demographic Adaptation in the Age of Transformers
Findings of EACL 2023
May, 2023
May, 2023
Demographic factors (e.g., gender or age) shape our language. Previous work showed that incorporating demographic factors can …
Know Your Audience: Do LLMs Adapt to Different Age and Education Levels?
April, 2023
Large language models (LLMs) offer a range of new possibilities, including adapting the text to different audiences and their reading …
Viewpoint: Artificial Intelligence Accidents Waiting to Happen?
JAIR
January, 2023
January, 2023
Artificial Intelligence (AI) is at a crucial point in its development: stable enough to be used in production systems, and increasingly …
Twitter-Demographer: A Flow-based Tool to Enrich Twitter Data
EMNLP 2023 Demos
December, 2022
December, 2022
Twitter data have become essential to Natural Language Processing (NLP) and social science research, driving various scientific …
Bridging Fairness and Environmental Sustainability in Natural Language Processing
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
December, 2022
December, 2022
Fairness and environmental impact are important research directions for the sustainable development of artificial intelligence. …
SocioProbe: What, When, and Where Language Models Learn about Sociodemographics
Proceedings of EMNLP 2022
December, 2022
December, 2022
Pre-trained language models (PLMs) have outperformed other NLP models on a wide range of tasks. Opting for a more thorough …
Data-Efficient Strategies for Expanding Hate Speech Detection into Under-Resourced Languages
2022 Conference on Empirical Methods in Natural Language Processing (EMNLP 2022)
October, 2022
October, 2022
Hate speech is a global phenomenon, but most hate speech datasets so far focus on English-language content. This hinders the …
Is It Worth the (Environmental) Cost? Limited Evidence for the Benefits of Diachronic Continuous Training
arXiv preprint arXiv:2210.07365
October, 2022
October, 2022
Language is constantly changing and evolving, leaving language models to quickly become outdated, both factually and linguistically. …
Welcome to the Modern World of Pronouns: Identity-Inclusive Natural Language Processing beyond Gender
Proceedings of the 29th International Conference on Computational Linguistics, COLING
October, 2022
October, 2022
The world of pronouns is changing – from a closed word class with few members to an open set of terms to reflect identities. However, …
Guiding the Release of Safer E2E Conversational AI through Value Sensitive Design
Proceedings of the 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue
September, 2022
September, 2022
Over the last several years, end-to-end neural conversational agents have vastly improved their ability to carry unrestricted, …
Hard and Soft Evaluation of NLP models with BOOtSTrap SAmpling - BooStSa
Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics: System Demonstrations
May, 2022
May, 2022
Natural Language Processing (NLP) ‘s applied nature makes it necessary to select the most effective and robust models. Producing …
Benchmarking Post-Hoc Interpretability Approaches for Transformer-based Misogyny Detection
First Workshop on Efficient Benchmarking in NLP at ACL 2022
April, 2022
April, 2022
Transformer-based Natural Language Processing models have become the standard for hate speech detection. However, the unconscious use …
Measuring Harmful Sentence Completion in Language Models for LGBTQIA+ Individuals
Second Workshop on Language Technology for Equality, Diversity and Inclusion at ACL 2022
April, 2022
April, 2022
Current language technology is ubiquitous and directly influences individuals' lives worldwide. Given the recent trend in AI on …
Pipelines for Social Bias Testing of Large Language Models
First Workshop on Challenges & Perspectives in Creating Large Language Models at ACL 2022
April, 2022
April, 2022
The maturity level of language models is now at a stage in which many companies rely on them to solve various tasks. However, while …
Two Contrasting Data Annotation Paradigms for Subjective NLP Tasks
Proceedings of NAACL-HLT 2022
April, 2022
April, 2022
Labelled data is the foundation of most natural language processing tasks. However, labelling data is difficult and there often are …
XLM-EMO: Multilingual Emotion Prediction in Social Media Text
12th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis at EACL 2021
April, 2022
April, 2022
Detecting emotion in text allows social and computational scientists to study how people behave and react to online events. However, …
Fair and Argumentative Language Modeling for Computational Argumentation
ACL
April, 2022
April, 2022
Although much work in NLP has focused on measuring and mitigating stereotypical bias in semantic spaces, research addressing bias in …
Entropy-based Attention Regularization Frees Unintended Bias Mitigation from Lists
Findings of the Association for Computational Linguistics: ACL2022 (Forthcoming)
March, 2022
March, 2022
Natural Language Processing (NLP) models risk overfitting to specific terms in the training data, thereby reducing their performance, …
SAFETYKIT: First Aid for Measuring Safety in Open-domain Conversational Systems
Findings of the Association for Computational Linguistics: ACL2022
March, 2022
March, 2022
The social impact of natural language processing and its applications has received increasing attention. In this position paper, we …
Five sources of bias in natural language processing
Language and Linguistics Compass
August, 2021
August, 2021
Recently, there has been an increased interest in demographically grounded bias in natural language processing (NLP) applications. Much …
Exposing the limits of Zero-shot Cross-lingual Hate Speech Detection
59th Annual Meeting of the Association for Computational Linguistics
August, 2021
August, 2021
Reducing and counter-acting hate speech on Social Media is a significant concern. Most of the proposed automatic methods are conducted …
HONEST: Measuring Hurtful Sentence Completion in Language Models
2021 Annual Conference of the North American Chapter of the Association for Computational Linguistics
June, 2021
June, 2021
Language models have revolutionized the field of NLP. However, language models capture and proliferate hurtful stereotypes, especially …
The Importance of Modeling Social Factors of Language: Theory and Practice
2021 Annual Conference of the North American Chapter of the Association for Computational Linguistics
June, 2021
June, 2021
Natural language processing (NLP) applications are now more powerful and ubiquitous than ever before. With rapidly developing (neural) …
