Indira Sen

Hi! I am a computer scientist and junior faculty at the Chair of Data Science for the Social and Economic Sciences in the University of Mannheim, Germany.

My research lies at the intersection of Natural Language Processing (NLP) and Computational Social Science. I investigate the social knowledge, values, and perspectives embedded in generative AI systems, e.g., political values in Large Language Models (LLMs) and their ability to assist and simulate different groups of people. Finally, I design methods for assessing social values in LLMs. By interrogating and improving social knowledge in these models, I hope to build language technology that supports and reflects diverse user bases and populations.

Before Mannheim, I was a postdoc at the University of Konstanz in the Department of Politics and Public Policy. I did my PhD at RWTH-Aachen and GESIS, Leibniz Institute for the Social Sciences, advised by Claudia Wagner and Wil van der Aalst. My doctoral research focused on developing valid and robust measurements of social constructs from digital trace data.

Before moving to Germany for my doctoral studies, I did my Master’s at IIIT Delhi and my Bachelor’s at Vellore Institute of Technology, both in India.

News

• Oct’26: Attending EMNLP in Budapest and presenting a poster at the NLP4PI workshop on investigating how LLMs answer questions about democracy
• Oct’26: Participating in a panel discussion on “Diversity and Gender in Research” at the University of Mannheim
• Sept’26: Gave an invited lecture as a part of the Lecture Series on AI & Survey Science. My lecture was titled “Is AI Friend or Foe to Survey-based Data Collection?”
• Fall-Winter Semester’26: teaching the 2nd iteration of “Large Language Models for the Economic and Social Sciences”
• I’m one of the program chairs of ICWSM’27!
• July’26: We organized a tutorial at IC2S2’26 on “Simulating Human Survey Responses with Large Language Models: Potentials and Pitfalls”
• June’26: Gave an invited talk at the University of Innsbruck’s Digital Science Center on “Measuring Political Values in Large Language Models”
• June’26: We organized a workshop at the University of Mannheim on “Shaping the Future of AI-Enabled Surveys”. Thanks to the Fritz-Thyssen Foundation, the Mannheim Centre for European Social Research, and the Mannheim Center for Data Science for supporting the workshop!