Metraipolis › archive › dc_cd754adb9e18

Induction and Inquiry via Probabilistic Reasoning over Language and Code — arXiv:2609.01815

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.01815
License (as archived)
CC0-1.0
Archived
2026-10-17T03:25:31.987Z
SHA-256 of the archived text
3f518d779217723d86403befd4c94b5c5ef8a4edc793a9e2676133be250b56f4
Stable URL of this page
https://metraipolis.com/documento/dc_cd754adb9e18

How to cite

Induction and Inquiry via Probabilistic Reasoning over Language and Code — arXiv:2609.01815 — arXiv (en), https://arxiv.org/abs/2609.01815 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cd754adb9e18 (Metraipolis archive dc_cd754adb9e18, sha256 3f518d779217723d86403befd4c94b5c5ef8a4edc793a9e2676133be250b56f4)
BibTeX
@misc{dccd754adb9e18,
  title = {Induction and Inquiry via Probabilistic Reasoning over Language and Code — arXiv:2609.01815},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.01815},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cd754adb9e18; sha256 3f518d779217723d86403befd4c94b5c5ef8a4edc793a9e2676133be250b56f4},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01815. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

How humans grow and maintain abstract knowledge from the sparse, streaming noisy data of experience is a longstanding challenge in cognitive science. Any computational account must satisfy at least three desiderata: It must be (1) data-efficient and compute-efficient, (2) capture gradations of uncertainty to support intelligent inquiry and information gathering, and (3) be flexible enough to mentally represent the endless range of concepts people can learn and think about. Here we introduce a computational model that captures these three properties, by encoding symbolic knowledge as mental programs that combine natural language with source code, and sequentially inferring mental programs using LLM-guided Bayesian learning algorithms. Across a range of behavioral studies this model successfully reproduces quantitative signatures of human inductive learning and active inquiry, such as anchoring, garden-pathing, and other effects. In contrast, pure LLMs and classic Bayesian models either fail at the underlying task, or do not reproduce human behavior, or succeed only at exorbitant computational cost. These results suggest that one way humans continually grow their knowledge is by mentally representing many hypotheses spanning language-like and program-like representations, then revising those hypotheses to approximate Bayesian updates, while a bottom-up neural mechanism (an LLM) makes inference both tractable and learnable.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-03 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.01815 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T03:13:12.400Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-03?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).