Dirk Hovy

Latest

• Divine LLaMAs: Bias, Stereotypes, Stigmatization, and Emotion Representation of Religion in Large Language Models
• Compromesso! Italian Many-Shot Jailbreaks Undermine the Safety of Large Language Models
• My Answer is C: First-Token Probabilities Do Not Match Text Answers in Instruction-Tuned Language Models
• Political Compass or Spinning Arrow? Towards More Meaningful Evaluations for Values and Opinions in Large Language Models
• XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Models
• DADIT: A Dataset for Demographic Classification of Italian Twitter Users and a Comparison of Prediction Methods
• Beyond Flesch-Kincaid: Prompt-based Metrics Improve Difficulty Classification of Educational Texts
• SafetyPrompts: a Systematic Review of Open Datasets for Evaluating and Improving Large Language Model Safety
• Angry Men, Sad Women: Large Language Models Reflect Gendered Stereotypes in Emotion Attribution
• Conversations as a Source for Teaching Scientific Concepts at Different Education Levels
• Emotion Analysis in NLP: Trends, Gaps and Roadmap for Future Directions
• Classist Tools: Social Class Correlates with Performance in NLP
• Impoverished Language Technology: The Lack of (Social) Class in NLP
• Wisdom of Instruction-Tuned Language Model Crowds. Exploring Model Label Variation
• MilaNLP at SemEval-2023 Task 10: Ensembling Domain-Adapted and Regularized Pretrained Language Models for Robust Sexism Detection
• Respectful or Toxic? Using Zero-Shot Learning with Language Models to Detect Hate Speech
• Temporal and Second Language Influence on Intra-Annotator Agreement and Stability in Hate Speech Labelling
• The Ecological Fallacy in Annotation: Modeling Human Label Variation goes beyond Sociodemographics
• The State of Profanity Obfuscation in Natural Language Processing Scientific Publications
• What about ''em''? How Commercial Machine Translation Fails to Handle (Neo-)Pronouns
• What about ''em''? How Commercial Machine Translation Fails to Handle (Neo-)Pronouns
• Leveraging Social Interactions to Detect Misinformation on Social Media
• Proceedings of the First Workshop on Cross-Cultural Considerations in NLP (C3NLP)
• Can Demographic Factors Improve Text Classification? Revisiting Demographic Adaptation in the Age of Transformers
• Know Your Audience: Do LLMs Adapt to Different Age and Education Levels?
• Beyond Digital 'Echo Chambers': The Role of Viewpoint Diversity in Political Discussion
• Viewpoint: Artificial Intelligence Accidents Waiting to Happen?
• It's Not Just Hate: A Multi-Dimensional Perspective on Detecting Harmful Speech Online
• Twitter-Demographer: A Flow-based Tool to Enrich Twitter Data
• Bridging Fairness and Environmental Sustainability in Natural Language Processing
• SocioProbe: What, When, and Where Language Models Learn about Sociodemographics
• Data-Efficient Strategies for Expanding Hate Speech Detection into Under-Resourced Languages
• Is It Worth the (Environmental) Cost? Limited Evidence for the Benefits of Diachronic Continuous Training
• Welcome to the Modern World of Pronouns: Identity-Inclusive Natural Language Processing beyond Gender
• Guiding the Release of Safer E2E Conversational AI through Value Sensitive Design
• Hard and Soft Evaluation of NLP models with BOOtSTrap SAmpling - BooStSa
• Language Invariant Properties in Natural Language Processing
• Benchmarking Post-Hoc Interpretability Approaches for Transformer-based Misogyny Detection
• Measuring Harmful Sentence Completion in Language Models for LGBTQIA+ Individuals
• Pipelines for Social Bias Testing of Large Language Models
• Two Contrasting Data Annotation Paradigms for Subjective NLP Tasks
• XLM-EMO: Multilingual Emotion Prediction in Social Media Text
• Entropy-based Attention Regularization Frees Unintended Bias Mitigation from Lists
• SAFETYKIT: First Aid for Measuring Safety in Open-domain Conversational Systems
• Text Analysis in Python for Social Scientists – Prediction and Classification
• Learning from Disagreement: A Survey
• Five sources of bias in natural language processing
• On the Gap between Adoption and Understanding in NLP
• Pre-training is a Hot Topic: Contextualized Document Embeddings Improve Topic Coherence
• 'We will Reduce Taxes' - Identifying Election Pledges with Language Models
• HONEST: Measuring Hurtful Sentence Completion in Language Models
• The Importance of Modeling Social Factors of Language: Theory and Practice
• FEEL-IT: Emotion and Sentiment Classification for the Italian Language
• MilaNLP @ WASSA: Does BERT Feel Sad When You Cry?
• Universal Joy A Data Set and Results for Classifying Emotions Across Languages
• BERTective: Language Models and Contextual Information for Deception Detection
• Cross-lingual Contextualized Topic Models with Zero-shot Learning
• Text Analysis in Python for Social Scientists – Discovery and Exploration
• Predictive Biases in Natural Language Processing Models: A Conceptual Framework and Overview
• “You Sound Just Like Your Father” Commercial Machine Translation Systems Include Stylistic Biases
• Visualizing Regional Language Variation Across Europe on Twitter
• Helpful or Hierarchical? Predicting the Communicative Strategies of Chat Participants, and their Impact on Success
• What the [MASK]? Making Sense of Language-Specific BERT Models
• A Case for Soft Loss Functions
• Dense Node Representation for Geolocation
• Geolocation with Attention-Based Multitask Learning Models
• Hey Siri. Ok Google. Alexa: A topic modeling of user reviews for smart speakers
• Identifying Linguistic Areas for Geolocation
• Women’s Syntactic Resilience and Men’s Grammatical Luck: Gender-Bias in Part-of-Speech Tagging and Dependency Parsing
• Peer networks and entrepreneurship: A Pan-African RCT
• Increasing In-Class Similarity by Retrofitting Embeddings with Demographic Information
• Capturing Regional Variation with Distributed Place Representations and Geographic Retrofitting
• Comparing Bayesian Models of Annotation
• Predicting News Headline Popularity with Syntactic and Semantic Knowledge Using Multi-Task Learning
• The Social and the Neural Network: How to Make Natural Language Processing about People again