Hypothesis Generation and Inductive Inference in Children and Language Models
Jeffrey Qin, Wasu Top Piriyakulkij, Zhuangfei Gao, Mia Radovanovic, Jessica Sommerville, Kevin Ellis, Marta Kryven · 2026
Read Hypothesis Generation and Inductive Inference in Children and Language Models by Jeffrey Qin, Wasu Top Piriyakulkij, Zhuangfei Gao, Mia Radovanovic, Jessica Sommerville, Kevin Ellis, Marta Kryven (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-24528.txt.
Description
Jeffrey Qin and colleagues compare how children and language-model agents form hypotheses under uncertainty, using a Box Task where players infer a cause. Children's behavior fits weighing evidence reliability plus forming hypotheses on the fly, explaining evidence seeking and why task success and rule learning come apart. Language-model agents mirror children's answers to reliability and observability shifts, discounting weak evidence. Yet the models over-observe and over-comply more than children, hinting at different costs behind adaptation.
Catalog record
| CREATOR | Jeffrey Qin, Wasu Top Piriyakulkij, Zhuangfei Gao, Mia Radovanovic, Jessica Sommerville, Kevin Ellis, Marta Kryven |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 72039 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.24528