Research Groups

Madison NLP brings together researchers working on diverse topics in natural language processing and artificial intelligence, from language and multimodal models to social and cognitive aspects of AI. Explore our research areas below.

Hu Lab

The Hu Lab focuses on building robust and adaptive AI systems that can reason effectively, operate safely in dynamic environments, and communicate across languages and cultures. We are particularly interested in understanding and improving large language models (LLMs), adapting them to knowledge-intensive reasoning tasks, and aligning them with the needs of diverse users.

Our current research spans four main areas:

• Multilingual NLP and Cultures: Democratizing inclusive LLMs across languages and cultures while enhancing pluralistic alignment.
• Efficient Multimodal LLMs: Developing efficient multimodal LLMs for high-stakes applications in education and healthcare.
• Cognitive Language Agents: Unifying cognitive concepts and languages to enable complex reasoning, planning and human behavior simulation.
• Evaluation and Mechanistic Interpretation: Exploring robust methods for evaluating and interpreting black-box foundation models.

SLAI Lab

The Social Language & AI (SLAI) lab at UW–Madison studies language and AI through social and human-centered perspectives. Our research brings together natural language processing, computational social science, and responsible AI to better understand how language technologies interact with people, communities, and culture.

Our research interests include:

• Human-Centered and Responsible AI: Understanding the societal impacts, safety, and fairness of language models and developing methods that better reflect diverse human needs and experiences.
• Computational Social Science: Using NLP and language models to study social and cultural phenomena across domains such as media, literature, and education.
• Language, Communities, and Culture: Studying how language varies across social groups and communities, and developing more culturally aware and inclusive NLP systems.
• Human–AI Interaction: Evaluating and improving how people communicate, collaborate, and interact with language models.

Jiang Lab

The Jiang Lab studies human-centered and trustworthy language technologies, with a particular focus on how AI systems perceive, reason about, and communicate information across diverse human and social contexts. Our research develops algorithms and benchmarks for understanding interactions between people and language or multimodal models, with the goal of enabling more reliable and effective human–AI and cross-community communication.

Our research interests include:

• Mental Modeling in LLMs and VLMs: Understanding how large language models and vision-language models represent and reason about people, perspectives, beliefs, and communicative contexts.
• Responsibility to Social Context: Developing and evaluating language technologies that respond appropriately to the social contexts in which communication takes place.
• Adaptability to Human-Centered Diversity: Improving the ability of AI systems to adapt to human diversity across a wide range of communication scenarios.