Chemical reasoning language models are expected to produce faithful chain-of-thought (CoT) explanations when answering chemistry tasks, but across four model families and twelve tasks, hallucinations are widespread and largely independent of answer correctness. Attribution analyses reveal that these models use a shared scratchpad function: Chem‑R and ether‑0 rely on fragmented SMILES drafts, while ChemDFM‑R emphasizes scaffold, positional, and naming cues. Perturbing Chem‑R’s SMILES sketches degrades generation, indicating that structural drafts can be causally load‑bearing even when verbal structural claims are largely inert.
arXiv:2602. 07075v5 Announce Type: replace-cross Abstract: Current chemical large language models (LLMs) predominantly rely on explicit Chain-of-Thought (CoT) to solve complex reasoning problems.
arXiv:2606. 03660v1 Announce Type: new Abstract: Large language models are increasingly used as chemistry assistants, yet most chemistry benchmarks still score only final answers.
arXiv:2606. 03057v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used for molecular tasks, but it remains unclear which molecular representation to use.
oMeBench is a large-scale, expert-curated benchmark designed to evaluate large language models (LLMs) on organic mechanism reasoning. It contains over 10,000 annotated mechanistic steps, including reaction type labels, intermediate structures, and difficulty ratings, and introduces the oMeS scoring framework to assess logical consistency and chemical structural similarity. Evaluation shows that while current LLMs display promising chemical intuition, they often fail to produce correct and consistent multi-step reasoning, though prompting and fine-tuning can bring smaller models up to the level of closed‑source frontier models.
ChemVTS-Bench is a domain-authentic benchmark that evaluates Visual‑Textual‑Symbolic reasoning in multimodal large language models for chemistry. It presents diverse chemical problems—organic molecules, inorganic materials, and 3D crystal structures—in three input modes: visual-only, visual‑text hybrid, and SMILES-based symbolic. The benchmark includes an automated agent workflow for inference, answer verification, and failure diagnosis, and shows that visual-only inputs and structural chemistry remain challenging for current models.
arXiv:2607. 12771v1 Announce Type: new Abstract: Reaction mechanisms consist of the step-by-step sequences of elementary reactions that explain chemical transformations.
arXiv:2608. 10480v1 Announce Type: new Abstract: Large language models (LLMs) are widely applied across chemical tasks, such as molecular property prediction, which underpins drug discovery.
arXiv:2512.00729v2 Announce Type: replace Abstract: Motivated by the observed human-like behaviours in Large Reasoning Models (LRMs), this paper introduces a comprehensive taxonomy to characterise at...
Large language models (LLMs) are widely applied across chemical tasks, such as molecular property prediction, which underpins drug discovery. Molecular LLMs represent a molecule through several modalities, notably a 1D SMILES sequence or a 2D molecular graph.
Large language models offer a promising foundation for chemical reasoning, bringing together chemical knowledge and multistep problem solving. Chemical intuition can provide an initial sense of plausi...
arXiv:2607. 02140v1 Announce Type: new Abstract: Chemical language models (CLMs) are trained with linearized representations such as SMILES, yet it remains unclear which chemically meaningful substructures they encode.
Latent JEPA is a new framework that trains continuous latent thoughts to anticipate informative aspects of future solutions in chemical reasoning, without verbalizing every intermediate step. It combines autoregressive learning with joint-embedding prediction of one or more future views, using textual and molecular prediction objectives that link latent thoughts to subsequent reasoning and molecular outcomes. Experiments on ChemCoTBench demonstrate improvements in molecular optimization, editing, and reaction metrics, and representation analyses show that future prediction makes latent thoughts more informative about molecular outcomes and better aligned with chemical structure.