Dirk Hovy · page 2
Papers
- The AI Gap: How Socioeconomic Status Affects Language Technology Interactions (2025 · 20 citations)
Socioeconomic status (SES) shapes how people communicate, but earlier studies linking SES to language technology relied on proxy measures or synthetic text. - IssueBench: Millions of Realistic Prompts for Measuring Issue Bias in LLM Writing Assistance (2026 · 16 citations)
LLMs are widely used as writing assistants, and a model that consistently favors one side of a political issue could gradually shift what users and their readers believe. - Know Your Audience: Do LLMs Adapt to Different Age and Education Levels? (2023 · 13 citations)
The paper asks whether large language models can tailor their answers to readers of different ages and education levels, which matters as these models enter classrooms. - Socially Aware Language Technologies: Perspectives and Practices (2025 · 12 citations)
This perspective essay from the "Last Words" column of Computational Linguistics argues that many recurring weaknesses of modern NLP, including biased outputs, toxicity, fragile user trust, uneven performance across groups and difficult evaluation, have a single root. - PATS: Personality-Aware Teaching Strategies with Large Language Model Tutors (2026 · 1 citation)
LLM tutors usually treat all learners alike, even though pedagogical research suggests that the best teaching strategy depends on a student's personality. - The Pluralistic Moral Gap: Understanding Moral Judgment and Value Differences between Humans and Large Language Models (2026 · 1 citation)
People increasingly ask LLMs for moral advice, but earlier evaluations mostly check agreement with the majority verdict. - ACID: On the Perception of Online Classism (2026)
In NLP, hate speech research has largely ignored discrimination based on socioeconomic status (SES). Few existing hate speech definitions even mention class. To fill this gap, the authors build ACID (Automatic Classism Identification Dataset), a corpus of more than 1.15 million English YouTube comments and Twitter… - Exploring Subjective Tasks in Farsi: A Survey Analysis and Evaluation of Language Models (2026)
Farsi has more than 127 million speakers and a large online presence, which would suggest a "middle-resource" status.
This page in Markdown → https://md.research.elamur.ai/free/en/authors/dirk-hovy/page/2/