Call for Papers
LLMSS Conference 2026
Large Language Models & the Social Sciences
October 14–16, 2026 · City University of Hong Kong
About the Conference
The inaugural Large Language Models and the Social Sciences Conference (LLMSS 2026) is organised by Talking to Machines — an international research initiative leveraging artificial intelligence in experiment design and implementation.
The conference brings together AI researchers, industry practitioners, government representatives, and social scientists to examine how large language models are reshaping research methodologies, data analysis, and theoretical frameworks across the social sciences.
Organising Institutions:
- Peking University — Analytics Lab for Global Risk Politics
- City University of Hong Kong — Computational Social Sciences Lab
- University of Oxford — Nuffield College Talking to Machines Initiative
An Open, International Conference
Large Language Models and the Social Sciences (LLMSS) Hong Kong Oct 14–16, 2026, is an open, international conference. We welcome submissions & attendance from researchers worldwide — regardless of institutional affiliation, nationality, or geopolitical context. Science advances through exchange!
Core Research Themes
Data Generation & Collection
- Language models in experimentation (AI agents as survey respondents)
- Adaptive experiment design through language model guidance
- Synthetic data generation and augmentation
- Dataset curation with automated labelling
LLM Applications in Social Science
- Text analysis and classification methods
- Causal inference from textual sources
- Analysis of interviews and open-ended responses
- Domain-specific applications in policy, economics, and law
LLM Development & Adaptation
- Model interpretability and explainability
- Fine-tuning and domain-specific customisation
- Prompt engineering strategies
- Evaluation and benchmarking protocols
Tools, Platforms & Infrastructure
- Research software libraries and frameworks
- LLM-driven research platforms
- Scalability and deployment solutions
- Reproducibility best practices
Ethics, Policy & Societal Impact
- Ethical challenges in research applications
- Fairness and bias mitigation
- Policy and governance implications
- Robustness and transparency standards
Multimodal & Emerging Methods
- Multimodal architecture innovations
- Cross-modal analysis (text, audio, video)
- Generative agents in social simulation
- Novel LLM applications in political science
Paper Submission
Key Dates
1 June 2026
Abstract Deadline
1 July 2026
Acceptance Notification
TBA
Camera-Ready Deadline
Full research papers and extended abstracts are accepted. Submissions are handled via the COMS conference submission system.
Registration
Registration opens August 2026
Details on registration fees and deadlines will be announced here.
Early Registration
€550
Until 31 July 2026
Regular Registration
€650
From 1 August 2026
Registration Deadline
15 August 2026
Accepted workshop participants receive dinner on the opening evening and lunch during the event. Travel and accommodation are not provided.
Cancellation Policy
- Before 15 August: Full refund minus €50 processing fee
- After 15 August or no-show: No refund
Conference Format
- Invited Keynote Talks: Talks by leading experts from academia and industry at the forefront of LLM research and applications.
- Research Paper and Poster Sessions: Presentations of accepted papers and poster displays for work-in-progress, with ample time for questions and discussion.
- Interactive Workshop Presentations: Seminar-style sessions where researchers can demo tools, share data, or conduct mini-tutorials on specialised methods.
- Methodological Demonstrations: Live demonstrations of new software, libraries, or experimental techniques relevant to LLMs and social science.
- Structured Feedback & Discussion: Dedicated discussant feedback for each presented paper and open-floor discussions to provide in-depth, constructive critique and foster collaboration.
Venue & Travel
City University of Hong Kong (CityU)
Tat Chee Avenue, Kowloon Tong, Hong Kong
The campus features modern, fully accessible facilities with direct MTR access via Kowloon Tong Station. It is adjacent to the Festival Walk shopping mall and is 30–40 minutes from Hong Kong International Airport.
October temperatures in Hong Kong typically range from 26–29°C (79–84°F), making it an ideal time for sightseeing. Nearby attractions include Victoria Peak, Kowloon Walled City Park, Mong Kok Markets, and the Tsim Sha Tsui Promenade.
Accommodation
Participants arrange their own accommodation. Early booking is recommended as October is peak season in Hong Kong.
About Talking to Machines
Talking to Machines is a project leveraging artificial intelligence in experiment design and implementation. It represents an international collaboration of academics and industry researchers across diverse institutions worldwide. Research findings have been published in leading peer-reviewed social science journals.
The project is directed by Sonja Vogt (HEC University of Lausanne) and Ray Duch (Nuffield College, University of Oxford), with primary funding from the Swiss National Science Foundation.
The main objective is to advance social science research through innovations that harness recent breakthroughs in AI.
Our Focus:
- Incorporating synthetic personas into experimental design
- Developing LLM models for crafting video intervention treatments
- Creating AI-enhanced tools reflecting cultural diversity in Global South research contexts
- Characterising treatment effect heterogeneity across populations
- Developing open-source platforms for research accessibility
Research Themes:
- AI–human subject interactions
- Digital twin technologies
- LLM-guided RCT data collection
- Adaptive experiments
- AI public opinion polling
- Video treatment design
Contact
For submission system issues, logistics questions, or conference scope enquiries, please contact:
Melanie Sawers — Conference Coordinator
melanie.sawers@nuffield.ox.ac.uk