IC²S² Tutorial: Potentials and Pitfalls of Simulating Human Survey Responses with LLMs
krisgligoric · x · 2026-07-29
At the 12th International Conference on Computational Social Science (IC²S²), researchers hosted a tutorial on simulating human survey responses using Large Language Models. The session focused on using 'silicon samples' to complement or extend human data, covering applications like survey pretesting, hybrid designs combining human and simulated respondents, and missing-data imputation.
The tutorial also addressed the new methodological choices and risks introduced by this approach, introducing the QSTN framework. This tool aims to structure simulations and support transparent evaluation across design alternatives, helping participants systematically explore how design decisions like persona construction and prompting strategies impact results.
Related event: IC2S2 Tutorial Explores Simulating Human Survey Responses with LLMs(2 posts)→
More from Research
- A KLS conjecture breakthrough links convex geometry, Monge–Ampère PDEs and ChatGPT — lihua_lei_stat · 2026-07-29
- Stream3D turns frozen 3D generators into streaming models with bounded memory — pliang279 · 2026-07-29
- Formal methods researchers discuss how to respond to AI progress at FLoC — swarat · 2026-07-29
- Hugging Face argues there is no truly tokenization-free LLM — cjmaddison · 2026-07-29
- PlayCanvas Update: Volumetric Fog Now Lit by Clustered Lights — willeastcott · 2026-07-29
- OKLS brings KL-optimal Shampoo to language model training with 1.45× parameter efficiency — aryaman2020 · 2026-07-29