6th Workshop on Computational Linguistics for the Political and Social Sciences (CPSS)

Call for papers

Special Theme: Human Perspectives and LLMs

We specifically invite submissions on this year’s special theme, focusing on the highly debated issue of simulation of human perspectives with LLMs. We are especially interested in papers addressing issues related to:

• Why: Why would we even want to simulate perspectives? What are the ethical boundaries that should not be crossed, even though we (seem to) have now a technology that can do “everything”?

• Who: Whose perspectives do we find (or will not find) in LLMs? How well do we do beyond English, or beyond WEIRD (Western, Educated, Industrialized, Rich, and Democratic)?

• What: What types of data are needed for this type of research? What are the constraints that these types of data impose?

• Where: The practical bottlenecks that determine where this type of research can be conducted (i.e., budget for compute, annotation) shape the community of the researchers that can work on this topic. Does this impact the diversity of research in this domain?

• How: What are the methodologically sound choices to make? What is the boundary between LLM positioning and bias? How about reproducibility?

General topics

In addition to the special theme, we also welcome submissions on the following general topics to CPSS:

• Modeling complex social constructs (e.g. populism, polarization, identity) with NLP methods

• Political and social bias in language models

• Modeling political communication with NLP (e.g. topic classification, position measurement)

• Mining policy debates from heterogeneous textual sources

• Methodological insights in interdisciplinary collaboration: workflows, challenges, best practices

• NLP support to understand and support democratic decision making

• Resources and tools for Political/Social Science research

• Validation of results beyond the train-dev-test paradigm of NLP and data science.

• Data quality in human and synthetic data

• Data leakage and contamination, especially in LLMs

• New ways to collect data, i.e., dataset donation

• Open science and reproducibility within the CPSS research

• and many more ...

🆕 PhD Forum

CPSS will feature a dedicated PhD Forum inviting submissions from doctoral researchers who wish to present their thesis topic. We warmly welcome contributions from PhD students at all stages of their doctorate, from those who are just beginning to develop their project to those with more advanced work in progress. 

The forum is designed to give early-career scholars the opportunity to discuss their projects with a broader audience, receive constructive feedback at different stages of their PhD, and connect with other researchers and PhD students from their own as well as related disciplines. Participants may choose whether they would like feedback primarily on the social/political science dimensions of their work or on its computational aspects. Accepted contributions will be presented both as spotlight talks and as posters in a dedicated PhD Forum session. 

General submission (archival and non-archival, July 10): https://openreview.net/group?id=GSCL.org/KONVENS/2026/Workshop/CPSS

Direct commitment for papers that are already peer reviewed on OpenReview (archival only, July 15): https://openreview.net/group?id=GSCL.org/KONVENS/2026/Workshop/CPSS_ARR_Commitment

• archival papers describing original and unpublished work (long papers: max. 8 pages, references/appendix excluded; short papers: max 4 pages, references/appendix excluded). Accepted papers will be published on the ACL anthology. For the submission format, refer to the KONVENS guidelines. When submitting, select "Full-paper submission".

• non-archival abstracts (1-page abstracts, references excluded) describing already published research or ongoing work. When submitting, select "Abstract submission".

• non-archival PhD project presentations (2-pages abstracts, references excluded) describing the idea, topic, and current state of the PhD project. When submitting, select "Abstract submission".

The three formats will meet the need of researchers from different communities, allowing the exchange of ideas in a "get to know each other" environment which we hope will foster future collaborations.

Submission format

Long/short archival paper submissions must use the same style template as KONVENS 2026, which is the ACL style template. Non-archival abstracts (including PhD project presentations) should be formatted according to this template and should be submitted as PDF. 

Long papers must not exceed eight (8) pages of content. Short papers and demonstration papers must not exceed four (4) pages of content. Non-archival abstracts must not exceed one (1) page. PhD project abstracts must not exceed two (2) pages. Limitations, Ethical Considerations, and References do not count against these limits. Supplementary material does also not count towards page limit and should be included in the appendix.

Every submission must be fully anonymized.

Important dates 

• Submission deadline (direct submission, archival and non archival):  July 10th

• Submission deadline ARR (archival only!):  July 15th

• Notification of acceptance:  August 7th

• Camera ready deadline: August 15th

• Workshop: September 14 (14.00-17.00) and 15 (9.00-12.00)

Organizers

• Dennis Assenmacher is a postdoctoral researcher in the Department of Computational Social Science at the Leibniz Institute for the Social Sciences (GESIS). His research focuses on harmful communication in online media (hate speech, abusive language, dehumanization, social bots), more specifically on developing state-of-the-art computational methods in the NLP domain to this harmful content.

• Agnieszka Faleńska is a computational linguist with a strong interest in the design, interpretability, and analysis of NLP architectures. She leads an independent research group in the Interchange Forum for Reflecting on Intelligent Systems (IRIS) at University of Stuttgart. She was a local co-organizer of the 16th ACM Web Science Conference 2024, co-organized ALGO 2014, and two editions of GeBNLP workshop in 2024 and 2025.

• Christopher Klamm is an interdisciplinary researcher with a background in Computer Science and Political Science. He is pursuing a PhD at the University of Mannheim, serving as a senior researcher at the Cologne Center of Comparative Politics, and is currently a research associate at Oxford’s Computational Political Science Group. His research combines computational social science and traditional social science methods.

• Gabriella Lapesa is a junior professor for Responsible Data Science and Machine Learning at HHU Düsseldorf and a team lead for Data Science methods at GESIS. He research interests include argument mining and deliberation, human label variation, and more broadly the interface of AL/LLMs and society, with the challenges, risks, and potential.

• Simone Paolo Ponzetto holds the chair of Information Systems III (Enterprise Data Analysis) at the University of Mannheim, where he leads the Natural Language Processing and Information Retrieval group. His research interests include research on text understanding and its interdisciplinary application in the Social Sciences and Humanities.

• Franziska Weeber is a PhD student at the Institute for Natural Language Processing at the University of Stuttgart and a guest researcher at the chair for Machine Learning at Humboldt Universität zu Berlin. Her research interests include how to measure and explain social and political opinions and biases in NLP models.

Program committee

Christian Arnold

Chung-hong Chan

Hongyu Chen

Clint Claessen

Agnese Daffara

Daryna Dementieva

Aditi Dutta

Leah von der Heyde

Zlata Kikteva

Andreas Küpfer

Verena Kunz

Jana Lasser

Maximilian Maurer

Sebastian Padó

Julia Romberg

Jens Rupprecht

Henning Wachsmuth

Tatjana Scheffler

Indira Sen

Melanie Siegel

Carina Silberer

Manfred Stede

Alexander Wuttke

Sukayna Younger-Khan

Sponsorship

We acknowledge the support of the Leibniz Institute for the Social Sciences (GESIS).