OpenAlex ·
Human Language Shapes the Emerging Physical Reasoning in Large Language Models
- Published
- Source
- OpenAlex
- Published in
- arXiv (Cornell University)
- DOI
- 10.48550/arxiv.2609.20666
- Original
- Read it at the publisher
- Topics
Abstract
Whether large language models (LLMs) acquire adequate representations of physical reality by predicting the next token remains unresolved. Molecular conformations provide a quantitative test: we asked 28 LLMs to rank 810 geometries of 27 molecules from atomic coordinates alone. Frontier LLMs approach the accuracy of specialized force fields, but their ability is bound to well-defined chemical concepts. They rank conformers with more hydrogen bonds as more stable, inverting the true order when the hydrogen-bonded conformer is less stable. On pairs with equal hydrogen-bond counts, they are most accurate for molecules whose explanations invoke hydrogen bonding, whereas molecules governed by ring strain remain difficult. Thus, the physical understanding emerging in LLMs inherits the structure and blind spots of human scientific discourse, a machine analogue of linguistic relativity.
The authors’ abstract, as arXiv lists it. arXiv’s listings, abstracts included, are free for anyone to reuse.
Read it on ReadSearch
Each opens after you create an account, and brings you back to this paper.
More in Materials Science
The Chern-Simons Function and the Quantum Hall Effect
Automated search for highly contextual Kochen-Specker proofs
Collective advantage from a minimal record in a quantum information engine
Enhanced thermal stability of SiGeSn by suppressing surface-mediated degradation
Anomalously enhanced lifetimes of low angular momentum Rydberg states in singly charged alkaline-earth metal ions
Get the few papers worth your time in Materials Science every morning.
ReadSearch reads the new papers in your field each day and brings you the few worth reading, each with its key finding in a sentence.