Rethinking Molecular Text Representations for LLMs: An Empirical Study

• URL: http://arxiv.org/abs/2606.03057v1
• Date: Tue, 02 Jun 2026 02:45:41 GMT
• Title: Rethinking Molecular Text Representations for LLMs: An Empirical Study
• Abstract summary: Large language models (LLMs) are increasingly used for molecular tasks, but it remains unclear which molecular representation to use.<n>We benchmark 16 LLMs across five model families, including reasoning and non-reasoning variants.<n>No single representation wins across tasks, though CML is the best, followed by MolJSON, InChI, and then canonical SMILES.
• Score: 1.7197939141715162
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models (LLMs) are increasingly used for molecular tasks, but it remains unclear which molecular representation to use. We present a systematic benchmark evaluating LLM molecular competence across nine representations and eight chemical tasks. We benchmark 16 LLMs across five model families, including reasoning and non-reasoning variants, chemistry-specialized LLMs, and closed frontier models. Performance is strongly representation-dependent and no single representation wins across tasks, though CML is the best, followed by MolJSON, InChI, and then canonical SMILES. Explicit structured text representations (CML and MolJSON) dominate structural tasks; IUPAC dominates semantic tasks, winning molecule retrieval for all 16 LLMs; and SMILES variants are rarely optimal despite their prevalence in pretraining. Chemistry-specialized models perform well with SMILES at the cost of large degradations with structured text representations, suggesting SMILES-only evaluation rewards specialization that does not generalize. Using LLM-as-a-judge, we find that IUPAC produces the highest fraction of correct molecule generations. A mechanistic study via tokenization audits, linear probes and attention shows that representations are encoded differently inside the model; for example, structured representations require higher attention across the molecular span. Our results argue against representation-invariant evaluation and motivate task-aware representation routing for LLM-based chemistry.

Related papers

• MolLingo: Molecule-Native Representations for LLM-Powered Scientific Agents [57.28652712579434]
  We present MolLingo, a multi-agent system that emulates the reasoning process of a chemist to automate molecular design.<n>MolLingo addresses this by coordinating a Literature Agent, a Chemist Agent, and an Orchestrator through a shared memory module, with each agent equipped with domain-specific tools.<n>As a case study in early-stage therapeutic design, MolLingo grounds the Chemist Agent's reasoning in binding site geometry and residue-level protein context.
  arXiv  Detail & Related papers  (2026-05-27T02:11:23Z)
• How well can off-the-shelf LLMs elucidate molecular structures from mass spectra using chain-of-thought reasoning? [51.286853421822705]
  Large language models (LLMs) have shown promise for reasoning-intensive scientific tasks, but their capability for chemical interpretation is still unclear.<n>We introduce a Chain-of-Thought (CoT) prompting framework and benchmark that evaluate how LLMs reason about mass spectral data to predict molecular structures.<n>Our evaluation across metrics of SMILES validity, formula consistency, and structural similarity reveals that while LLMs can produce syntactically valid and partially plausible structures, they fail to achieve chemical accuracy or link reasoning to correct molecular predictions.
  arXiv  Detail & Related papers  (2026-01-09T20:08:42Z)
• How to Make Large Language Models Generate 100% Valid Molecules? [82.89165081942201]
  Large language models (LLMs) can learn to perform a wide range of tasks from just a few examples.<n> generating valid molecules using representations like SMILES is challenging for LLMs in few-shot settings.<n>We introduce SmiSelf, a cross-chemical language framework for invalid SMILES correction.
  arXiv  Detail & Related papers  (2025-09-27T04:14:19Z)
• Improving Chemical Understanding of LLMs via SMILES Parsing [18.532188836688928]
  CLEANMOL is a novel framework that formulates SMILES parsing into a suite of clean and deterministic tasks.<n>We construct a molecular pretraining dataset with adaptive difficulty scoring and pre-train open-source LLMs on these tasks.<n>Our results show that CLEANMOL not only enhances structural comprehension but also achieves the best or competes with the baseline on the Mol-Instructions benchmark.
  arXiv  Detail & Related papers  (2025-05-22T07:54:39Z)
• ChemMLLM: Chemical Multimodal Large Language Model [52.95382215206681]
  We propose ChemMLLM, a unified chemical multimodal large language model for molecule understanding and generation.<n>Also, we design five multimodal tasks across text, molecular SMILES strings, and image, and curate the datasets.<n> Experimental results show that ChemMLLM achieves superior performance across all evaluated tasks.
  arXiv  Detail & Related papers  (2025-05-22T07:32:17Z)
• ChemEval: A Comprehensive Multi-Level Chemical Evaluation for Large Language Models [62.37850540570268]
  Existing benchmarks in this domain fail to adequately meet the specific requirements of chemical research professionals. ChemEval identifies 4 crucial progressive levels in chemistry, assessing 12 dimensions of LLMs across 42 distinct chemical tasks. Results show that while general LLMs excel in literature understanding and instruction following, they fall short in tasks demanding advanced chemical knowledge.
  arXiv  Detail & Related papers  (2024-09-21T02:50:43Z)
• MolX: Enhancing Large Language Models for Molecular Understanding With A Multi-Modal Extension [44.97089022713424]
  Large Language Models (LLMs) with their strong task-handling capabilities have shown remarkable advancements across a spectrum of fields.<n>This study seeks to enhance the ability of LLMs to comprehend molecules by equipping them with a multi-modal external module, termed MolX.<n>A hand-crafted molecular fingerprint is incorporated to leverage its embedded domain knowledge.
  arXiv  Detail & Related papers  (2024-06-10T20:25:18Z)
• Benchmarking Large Language Models for Molecule Prediction Tasks [7.067145619709089]
  Large Language Models (LLMs) stand at the forefront of a number of Natural Language Processing (NLP) tasks. This paper explores a fundamental question: Can LLMs effectively handle molecule prediction tasks? We identify several classification and regression prediction tasks across six standard molecule datasets. We compare their performance with existing Machine Learning (ML) models, which include text-based models and those specifically designed for analysing the geometric structure of molecules.
  arXiv  Detail & Related papers  (2024-03-08T05:59:56Z)
• Can Large Language Models Empower Molecular Property Prediction? [16.5246941211725]
  Molecular property prediction has gained significant attention due to its transformative potential in scientific disciplines. Recently, the rapid development of Large Language Models (LLMs) has revolutionized the field of NLP. In this work, we advance towards this objective through two perspectives: zero/few-shot molecular classification, and using the new explanations generated by LLMs as representations of molecules.
  arXiv  Detail & Related papers  (2023-07-14T16:06:42Z)
• Empowering Molecule Discovery for Molecule-Caption Translation with Large Language Models: A ChatGPT Perspective [53.300288393173204]
  Large Language Models (LLMs) have shown remarkable performance in various cross-modal tasks. In this work, we propose an In-context Few-Shot Molecule Learning paradigm for molecule-caption translation. We evaluate the effectiveness of MolReGPT on molecule-caption translation, including molecule understanding and text-based molecule generation.
  arXiv  Detail & Related papers  (2023-06-11T08:16:25Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.