Notation matters: cross-representation inconsistency in chemistry language models and its mechanistic origins
Digital Discovery · 2026 · 5 (11) · 4729-4738
DOI: 10.1039/d6dd00309e ↗Kurz gefasst
Fachartikel in der Zeitschrift Digital Discovery, 2026. Verfasst von Animesh Mishra. Die Arbeit nennt 18 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 3.0).
Abstract (Originalsprache)
Predictions from a chemistry language model shift by 1.69 log S on average when the same molecule is rewritten in another notation, affecting 88% of ESOL. Descriptor models are unaffected, locating the failure in string-based models.
Zitieren
APA
Mishra, A. (2026). Notation matters: cross-representation inconsistency in chemistry language models and its mechanistic origins. Digital Discovery, 5(11), 4729-4738. https://doi.org/10.1039/d6dd00309e
Vancouver
Mishra A. Notation matters: cross-representation inconsistency in chemistry language models and its mechanistic origins. Digital Discovery. 2026;5(11):4729-4738. doi:10.1039/d6dd00309e
BibTeX
@article{Mishra2026,
author = {Mishra, Animesh},
title = {Notation matters: cross-representation inconsistency in chemistry language models and its mechanistic origins},
journal = {Digital Discovery},
year = {2026},
volume = {5},
number = {11},
pages = {4729--4738},
doi = {10.1039/d6dd00309e}
}
EUVIMED