LLM research

    This research programme uses large language models to analyze political texts. The lead papers are “Using Large Language Models to Analyze Political Texts Through Natural Language Understanding” (American Journal of Political Science, 2026) and “Measuring issue salience for political parties using LLMs” (West European Politics, 2026), with related writing in the LSE Public Policy Review (2024) and Asian Management Insights (2025).

    Collaborators: Scott De Marchi, Conor Laver, Michael Laver and Jinshuai Ma.

    The work feeds into the quallmer package and the book AI-Assisted Content Analysis.

    Related publications

    1. 2026. “Measuring issue salience for political parties using LLMs.” West European Politics (with Michael Laver).

    2. 2026. “Using Large Language Models to Analyze Political Texts Through Natural Language Understanding.” American Journal of Political Science 70(4): 1589–1605 (with Scott De Marchi, Conor Laver, Michael Laver, and Jinshuai Ma).

    3. 2025. “Forging The Future: How AI is rewriting the rules of knowledge, expertise, and practice.” Asian Management Insights 12(2): 12–19.

    4. 2024. “AI and Data Science for Public Policy.” LSE Public Policy Review 3(3): 1–6.