#LLMSS2026
The notebooks from the three #LLMSS2026 workshops are now public.

Every notebook runs end to end with no network connection and no API key, so you can rerun and adapt them on your own laptop.

talking-to-machines.github.io/llmss-2026-w...
October 30, 2026 at 1:11 PM
What turned up at #LLMSS2026? Ray Duch’s wrap-up on where the field is putting its energy, and three threads that surfaced independently across panels.

200 people, 117 submissions, 138 institutions, 21 countries.

talkingtomachines.substack.com/p/what-turne...
What turned up
Three days in Hong Kong, 117 submissions, and a field that has started borrowing the right tools — the follow-up I promised.
talkingtomachines.substack.com
October 28, 2026 at 2:10 PM
That’s a wrap on the inaugural #LLMSS2026. Thank you to City University of Hong Kong, our speakers and everyone who came.

See you at LLMSS 2027!
October 16, 2026 at 3:07 PM
Keynote 4 at #LLMSS2026: Scott Hale (OII) on generative AI and access to quality information.

“You find the relationship-seeking model more engaging. That doesn’t change. But the gap between these two narrows significantly over four weeks.”
October 16, 2026 at 2:31 AM
Day 2 of #LLMSS2026: keynotes from Meng Tianguang (Tsinghua) and Issa Dahabreh (Harvard Chan), three roundtables and eight paper sessions.

Tomorrow: Scott Hale (Oxford Internet Institute)!
October 15, 2026 at 4:43 PM
Can AI-driven social research still produce human understanding?

Plenary roundtable at #LLMSS2026 with Jonathan Zhu (CityU), King-wa Fu (HKU) and Scott Hale (OII), chaired by Edmund Cheng.
October 15, 2026 at 4:51 AM
Keynote 2 at #LLMSS2026: Meng Tianguang (Tsinghua) on AI simulation and the black box of government responsiveness in China.

“For the central leader, the most important one is central–local alignment. It’s not stability.”
October 15, 2026 at 2:12 AM
Keynote 3 at #LLMSS2026: Issa Dahabreh (Harvard Chan) on using LLMs to make small experiments more efficient.

“In the social experiments, the LLMs did occasionally impressively well. I suspect this is because they bring to bear massive amounts of online information.”
October 15, 2026 at 7:16 AM
Day 1 of #LLMSS2026 at City University of Hong Kong: a keynote on simulating society, four parallel sessions and a roundtable on when simulations count as evidence.

Today: Meng Tianguang (Tsinghua) and Issa Dahabreh (Harvard Chan).
October 15, 2026 at 1:19 AM
We are open. The inaugural LLMSS conference begins today at City University of Hong Kong, with EPSA support. #LLMSS2026
October 14, 2026 at 5:51 AM
Keynote 1 at #LLMSS2026: Zhongyu Wei (Fudan) on SocioVerse 2.0, a longitudinal, dynamic framework for simulating society.

“Personalized LLMs are evolving from adapting to individuals toward understanding and simulating human behaviour”
October 14, 2026 at 6:56 AM
1 week left to submit 👇

Ray Duch (Talking to Machines) on LLMSS 2026 — LLMs & the Social Sciences Conference, 14–16 Oct, City University of Hong Kong. Hands-on workshops the day before, open to all.

📅 Abstracts due 7 Aug
🔗 coms.app/epsa-llm-26

#LLMSS2026 #AI #SocialScience
August 2, 2026 at 5:17 AM
Still time to submit for #LLMSS2026 at
City University of Hong Kong! We’re exploring how AI & LLMs are reshaping the social sciences—from political science to survey methodology & AI governance.
📅 Abstracts Due: 31 July 2026
🗓️ Conference: 14–16 Oct 2026
🔗 Submit: www.eps-academic.org/ai-conference
LLMSS 2026 — Large Language Models & the Social Sciences Conference | Talking to Machines
Open international conference on LLMs and social science research by Talking to Machines (T2M). October 14-16, City University of Hong Kong. Co-organised by Peking University, CityU & Oxford Nuffield....
www.eps-academic.org
July 20, 2026 at 9:10 PM