MolecularIQ: Characterizing Chemical Reasoning Capabilities Through Symbolic Verification on Molecular Graphs

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Bartmann, Christoph, Schimunek, Johannes, Ielanskyi, Mykyta, Seidl, Philipp, Klambauer, Günter, Luukkonen, Sohvi
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866914270976933888
author Bartmann, Christoph
Schimunek, Johannes
Ielanskyi, Mykyta
Seidl, Philipp
Klambauer, Günter
Luukkonen, Sohvi
author_facet Bartmann, Christoph
Schimunek, Johannes
Ielanskyi, Mykyta
Seidl, Philipp
Klambauer, Günter
Luukkonen, Sohvi
contents A molecule's properties are fundamentally determined by its composition and structure encoded in its molecular graph. Thus, reasoning about molecular properties requires the ability to parse and understand the molecular graph. Large Language Models (LLMs) are increasingly applied to chemistry, tackling tasks such as molecular name conversion, captioning, text-guided generation, and property or reaction prediction. Most existing benchmarks emphasize general chemical knowledge, rely on literature or surrogate labels that risk leakage or bias, or reduce evaluation to multiple-choice questions. We introduce MolecularIQ, a molecular structure reasoning benchmark focused exclusively on symbolically verifiable tasks. MolecularIQ enables fine-grained evaluation of reasoning over molecular graphs and reveals capability patterns that localize model failures to specific tasks and molecular structures. This provides actionable insights into the strengths and limitations of current chemistry LLMs and guides the development of models that reason faithfully over molecular structure.
format Preprint
id arxiv_https___arxiv_org_abs_2601_15279
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle MolecularIQ: Characterizing Chemical Reasoning Capabilities Through Symbolic Verification on Molecular Graphs
Bartmann, Christoph
Schimunek, Johannes
Ielanskyi, Mykyta
Seidl, Philipp
Klambauer, Günter
Luukkonen, Sohvi
Machine Learning
Artificial Intelligence
A molecule's properties are fundamentally determined by its composition and structure encoded in its molecular graph. Thus, reasoning about molecular properties requires the ability to parse and understand the molecular graph. Large Language Models (LLMs) are increasingly applied to chemistry, tackling tasks such as molecular name conversion, captioning, text-guided generation, and property or reaction prediction. Most existing benchmarks emphasize general chemical knowledge, rely on literature or surrogate labels that risk leakage or bias, or reduce evaluation to multiple-choice questions. We introduce MolecularIQ, a molecular structure reasoning benchmark focused exclusively on symbolically verifiable tasks. MolecularIQ enables fine-grained evaluation of reasoning over molecular graphs and reveals capability patterns that localize model failures to specific tasks and molecular structures. This provides actionable insights into the strengths and limitations of current chemistry LLMs and guides the development of models that reason faithfully over molecular structure.
title MolecularIQ: Characterizing Chemical Reasoning Capabilities Through Symbolic Verification on Molecular Graphs
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2601.15279