arxiv: 2602.04849 · v2 · submitted 2026-02-04 · ⚛️ physics.chem-ph · cs.AI· cs.MA

Recognition: no theorem link

El Agente Estructural: An Artificially Intelligent Molecular Editor

Changhyeok Choi , Yunheng Zou , Marcel M\"uller , Han Hao , Yeonghun Kang , Juan B. P\'erez-S\'anchez , Ignacio Gustin , Hanyong Xu

show 5 more authors

Andrew Wang Mohammad Ghazi Vakili Chris Crebolder Al\'an Aspuru-Guzik Varinia Bernales

Authors on Pith no claims yet

Pith reviewed 2026-05-16 06:49 UTC · model grok-4.3

classification ⚛️ physics.chem-ph cs.AIcs.MA

keywords molecular editorAI agentvision-language modelscomputational chemistrygeometry manipulationmultimodal AIautonomous modeling3D molecular editing

0 comments

The pith

El Agente Estructural is a multimodal AI agent that performs precise 3D molecular geometry edits using natural language and domain-informed tools.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces El Agente Estructural as an artificially intelligent molecular editor that mimics human experts in manipulating molecular systems in three dimensions. It combines a set of specialized geometry tools with vision-language models to enable direct control over atomic replacements, connectivity changes, and stereochemistry. This design avoids the need to rebuild entire molecular frameworks from scratch. Representative case studies demonstrate its use in site-selective functionalization, ligand binding and exchange, stereochemically controlled constructions, and generating structures from schematic reaction images. The agent is positioned as a component for larger autonomous quantum chemistry platforms.

Core claim

Through integration of domain-informed tools and vision-language models, El Agente Estructural enables chemically meaningful geometry manipulation in scenarios including site-selective functionalization, ligand binding, stereochemically controlled structure construction, and image-guided generation from reaction mechanisms, providing precise control over atomic connectivity and stereochemistry.

What carries the argument

Multimodal reasoning via vision-language models paired with specialized geometry-aware tools for interactive molecular editing.

Load-bearing premise

The combination of domain-informed tools and vision-language models will produce precise, error-free control over atomic connectivity and stereochemistry in complex cases without extensive human correction.

What would settle it

Observing whether the agent produces an accurate stereochemically correct structure from an ambiguous schematic reaction image without any manual correction.

read the original abstract

We present El Agente Estructural, a multimodal, natural-language-driven geometry-generation and manipulation agent for autonomous chemistry and molecular modelling. Unlike molecular generation or editing via generative models, Estructural mimics how human experts directly manipulate molecular systems in three dimensions by integrating a comprehensive set of domain-informed tools and vision-language models. This design enables precise control over atomic or functional group replacements, atomic connectivity, and stereochemistry without the need to rebuild extensive core molecular frameworks. Through a series of representative case studies, we demonstrate that Estructural enables chemically meaningful geometry manipulation across a wide range of real-world scenarios. These include site-selective functionalization, ligand binding, ligand exchange, stereochemically controlled structure construction, isomer interconversion, fragment-level structural analysis, image-guided generation of structures from schematic reaction mechanisms, and mechanism-driven geometry generation and modification. These examples illustrate how multimodal reasoning, when combined with specialized geometry-aware tools, supports interactive and context-aware molecular modelling beyond structure generation. Looking forward, the integration of Estructural into El Agente Quntur, an autonomous multi-agent quantum chemistry platform, enhances its capabilities by adding sophisticated tools for the generation and editing of three-dimensional structures.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

El Agente Estructural pairs vision-language models with chemistry-specific editing tools to handle natural-language 3D molecular changes, shown through case studies rather than benchmarks.

read the letter

The paper's main contribution is a multimodal agent that lets users describe molecular edits in plain language and have the system apply them directly in 3D space. It combines a vision-language model with a set of domain tools that manage replacements, connectivity, and stereochemistry without forcing a full rebuild of the structure each time. This setup is presented as closer to how human modelers actually work than pure generative approaches. The case studies walk through site-selective functionalization, ligand binding and exchange, stereocontrolled construction, isomer interconversion, and pulling structures from schematic reaction images. These examples are chemically plausible and illustrate the intended interactivity. The integration with the larger El Agente Quntur platform is a reasonable next step for embedding the editor in broader autonomous workflows. The architecture description is straightforward and the tool choices make sense for preserving geometry and bonds. The main gap is evaluation. Everything rests on selected examples with no success rates, failure analysis, or comparisons to existing editors or manual methods. The abstract talks about precise control, but without those numbers it is difficult to judge reliability on harder cases or ambiguous prompts. This is the kind of practical tool paper that computational chemists working on iterative design or reaction planning would find useful. It gives a working demonstration of interactive editing that goes beyond standard generation. I would send it for peer review. The system is clearly described and addresses a real workflow need, even if reviewers will want more systematic testing added.

Referee Report

0 major / 2 minor

Summary. The manuscript presents El Agente Estructural, a multimodal AI agent that integrates vision-language models with domain-informed geometry tools to enable natural-language-driven manipulation of molecular structures in 3D. It claims precise control over atomic connectivity, functional group replacements, and stereochemistry without rebuilding core frameworks, demonstrated via representative case studies covering site-selective functionalization, ligand binding/exchange, stereochemically controlled construction, isomer interconversion, fragment analysis, and image-guided generation from schematic reaction mechanisms. The system is further positioned for integration into the larger El Agente Quntur autonomous quantum chemistry platform.

Significance. If the case studies hold, the work provides a practical advance in interactive molecular modeling by offering an agent-based approach that directly mimics expert human 3D manipulation rather than depending on generative models alone. This could improve efficiency in structure editing tasks central to chemistry and drug design, with the multimodal tool-plus-VLM design offering a plausible route to context-aware interactivity. The framing as a deployable component within a broader autonomous platform adds forward-looking value for computational chemistry workflows.

minor comments (2)

The abstract and case-study descriptions would benefit from explicit mention of the total number of examples presented and the specific molecular systems used, to better contextualize the scope of the demonstrations.
Figure captions and any accompanying visuals should include clear labels for before/after structures, atom indices, and stereochemical descriptors to aid reader verification of the claimed manipulations.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their positive summary and recommendation of minor revision. No specific major comments were provided in the report, indicating that the core contributions and case studies are viewed favorably. We will address any minor points during revision.

Circularity Check

0 steps flagged

No significant circularity in tool demonstration

full rationale

The paper describes a multimodal software agent for molecular geometry editing that combines domain-informed tools with vision-language models. Its central claims rest on qualitative case studies illustrating capabilities such as site-selective functionalization, stereochemical control, and image-guided structure generation. No mathematical derivations, fitted parameters, equations, or self-referential predictions appear in the presented material; the work is explicitly framed as a system demonstration rather than a theoretical result derived from inputs. Consequently the derivation chain is self-contained with no reduction of outputs to inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The central claim rests on the assumption that existing vision-language models can be reliably combined with domain-specific geometry tools to achieve chemically accurate edits; no free parameters or new physical entities are introduced.

axioms (1)

domain assumption Vision-language models can correctly interpret molecular images and text instructions for geometry tasks
Invoked when describing integration of VLMs with editing tools

invented entities (1)

El Agente Estructural no independent evidence
purpose: Multimodal natural-language molecular geometry editor
New system presented in the paper

pith-pipeline@v0.9.0 · 5564 in / 1274 out tokens · 37342 ms · 2026-05-16T06:49:29.378529+00:00 · methodology

discussion (0)

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

ToolMol: Evolutionary Agentic Framework for Multi-objective Drug Discovery
cs.LG 2026-05 unverdicted novelty 6.0

ToolMol integrates evolutionary algorithms with agentic LLMs and precise RDKit tools to optimize multi-objective drug properties, yielding ligands with over 10% better predicted binding affinity and 35% gains in absol...
ToolMol: Evolutionary Agentic Framework for Multi-objective Drug Discovery
cs.LG 2026-05 unverdicted novelty 5.0

ToolMol is an evolutionary agentic framework that pairs multi-objective genetic algorithms with LLM tool-calling to generate drug-like ligands with over 10% better predicted binding affinity and 35% better ABFE scores...

Reference graph

Works this paper leans on

87 extracted references · 87 canonical work pages · cited by 1 Pith paper

[1]

& Liu, F

Houk, K. & Liu, F. Holy grails for computational organic chemistry and biochemistry.Acc. Chem. Res.50, 539–543 (2017)

work page 2017
[2]

K., Bligaard, T., Rossmeisl, J

Nørskov, J. K., Bligaard, T., Rossmeisl, J. & Christensen, C. H. Towards the computational design of solid catalysts.Nat. Chem.1, 37–46 (2009)

work page 2009
[3]

Huang, B., von Rudorff, G. F. & von Lilienfeld, O. A. The central role of density functional theory in the ai age. Science381, 170–175 (2023)

work page 2023
[4]

Mater.15, 1120–1127 (2016)

Gómez-Bombarelli, R.et al.Design of efficient molecular organic light-emitting diodes by a high-throughput virtual screening and experimental approach.Nat. Mater.15, 1120–1127 (2016)

work page 2016
[5]

F., Bonde, J., Chorkendorff, I

Greeley, J., Jaramillo, T. F., Bonde, J., Chorkendorff, I. & Nørskov, J. K. Computational high-throughput screening of electrocatalytic materials for hydrogen evolution.Nat. Mater.5, 909–913 (2006)

work page 2006
[6]

W.et al.Combining theory and experiment in electrocatalysis: Insights into materials design.Science 355, eaad4998 (2017)

Seh, Z. W.et al.Combining theory and experiment in electrocatalysis: Insights into materials design.Science 355, eaad4998 (2017)

work page 2017
[7]

H.et al.Progress in computational and machine-learning methods for heterogeneous small-molecule activation.Adv

Gu, G. H.et al.Progress in computational and machine-learning methods for heterogeneous small-molecule activation.Adv. Mater.32, 1907865 (2020)

work page 2020
[8]

Greeley, J.et al.Alloys of platinum and early transition metals as oxygen reduction electrocatalysts.Nat. Chem. 1, 552–556 (2009)

work page 2009
[9]

J., Bell, A

Garza, A. J., Bell, A. T. & Head-Gordon, M. Mechanism of CO2 reduction at copper surfaces: pathways to C2 products.ACS. Catal.8, 1490–1499 (2018)

work page 2018
[10]

A., Abild-Pedersen, F., Studt, F., Rossmeisl, J

Peterson, A. A., Abild-Pedersen, F., Studt, F., Rossmeisl, J. & Nørskov, J. K. How copper catalyzes the electroreduction of carbon dioxide into hydrocarbon fuels.Energy Environ. Sci.3, 1311–1315 (2010)

work page 2010
[11]

Zou, Y.et al.El agente: An autonomous agent for quantum chemistry.Matter8, 102263 (2025)

work page 2025
[12]

arXiv:2505.08195 [physics.comp-ph]

Hu, J.et al.Aitomia: Your intelligent assistant for ai-driven atomistic and quantum chemical simulations (2025). arXiv:2505.08195 [physics.comp-ph]

work page arXiv 2025
[13]

arXiv:2507.14267 [cs.AI]

Wang, Z.et al.Dreams: Density functional theory based research engine for agentic materials simulation (2025). arXiv:2507.14267 [cs.AI]

work page arXiv 2025
[14]

B.et al.El agente quntur: A research collaborator agent for quantum chemistry (2026)

Pérez-Sánchez, J. B.et al.El agente quntur: A research collaborator agent for quantum chemistry (2026)

work page 2026
[15]

P.et al.An open source chemical structure curation pipeline using rdkit.J

Bento, A. P.et al.An open source chemical structure curation pipeline using rdkit.J. Cheminform.12, 51 (2020)

work page 2020
[16]

M.et al.Open babel: An open chemical toolbox.J

O’Boyle, N. M.et al.Open babel: An open chemical toolbox.J. Cheminform.3, 33 (2011)

work page 2011
[17]

Sci.4, 268–276 (2018)

Gómez-Bombarelli, R.et al.Automatic Chemical Design Using a Data-Driven Continuous Representation of Molecules.ACS Cent. Sci.4, 268–276 (2018)

work page 2018
[18]

H., Kogej, T., Tyrchan, C

Segler, M. H., Kogej, T., Tyrchan, C. & Waller, M. P. Generating focused molecule libraries for drug discovery with recurrent neural networks.ACS Cent. Sci.4, 120–131 (2018)

work page 2018
[19]

Inf.37, 1700111 (2018)

Gupta, A.et al.Generative recurrent networks for de novo drug design.Mol. Inf.37, 1700111 (2018)

work page 2018
[20]

& Tropsha, A

Popova, M., Isayev, O. & Tropsha, A. Deep reinforcement learning for de novo drug design.Sci. Adv.4, eaap7885 (2018)

work page 2018
[21]

& Aspuru-Guzik, A

Sanchez-Lengeling, B. & Aspuru-Guzik, A. Inverse molecular design using machine learning: Generative models for matter engineering.Science361, 360–365 (2018)

work page 2018
[22]

O., Rupp, M

Ramakrishnan, R., Dral, P. O., Rupp, M. & Von Lilienfeld, O. A. Quantum chemistry structures and properties of 134 kilo molecules.Sci. data1, 1–7 (2014)

work page 2014
[23]

Krenn, M.et al.Selfies and the future of molecular string representations.Patterns3(2022)

work page 2022
[24]

D.et al.Avogadro: an advanced semantic chemical editor, visualization, and analysis platform.J

Hanwell, M. D.et al.Avogadro: an advanced semantic chemical editor, visualization, and analysis platform.J. Cheminform.4, 17 (2012)

work page 2012
[25]

Dennington, R., Keith, T. A. & Millam, J. M. Gaussview Version 6 (2019). Semichem Inc. Shawnee Mission KS

work page 2019
[26]

Iqmol: A molecular builder and visualization package

Gilbert, A. Iqmol: A molecular builder and visualization package. 28

work page
[27]

& Noordik, J

Schaftenaar, G. & Noordik, J. H. Molden: a pre-and post-processing program for molecular and electronic structures.J. Comput. Aided Mol. Des.14, 123–134 (2000)

work page 2000
[28]

version 1.8, build 682

Chemcraft - graphical software for visualization of quantum chemistry computations. version 1.8, build 682

work page
[29]

& Lai, L

Li, J. & Lai, L. SynCraft: Guiding Large Language Models to Predict Edit Sequences for Molecular Synthesizability Optimization (2025). arXiv:2512.20333 [cs.AI]

work page arXiv 2025
[30]

arXiv:2510.04704 [cond-mat]

Lv, T.et al.AtomWorld: A Benchmark for Evaluating Spatial Reasoning in Large Language Models on Crystalline Materials (2025). arXiv:2510.04704 [cond-mat]

work page arXiv 2025
[31]

& Griffiths, T

Sumers, T., Yao, S., Narasimhan, K. & Griffiths, T. Cognitive architectures for language agents.Transactions on Machine Learning Research(2023)

work page 2023
[32]

H.et al.The atomic simulation environment—a Python library for working with atoms.J

Larsen, A. H.et al.The atomic simulation environment—a Python library for working with atoms.J. Phys. Condens. Matter29, 273002 (2017)

work page 2017
[33]

& Tanaka, I

Togo, A., Shinohara, K. & Tanaka, I. Spglib: a software library for crystal symmetry search.Sci. Technol. Adv. Mater. Methods4, 2384822 (2024)

work page 2024
[34]

P.et al.Python materials genomics (pymatgen): A robust, open-source python library for materials analysis.Comput

Ong, S. P.et al.Python materials genomics (pymatgen): A robust, open-source python library for materials analysis.Comput. Mater. Sci.68, 314–319 (2013)

work page 2013
[35]

& Grimme, S

Bannwarth, C., Ehlert, S. & Grimme, S. GFN2-xTB - An Accurate and Broadly Parametrized Self-Consistent Tight-Binding Quantum Chemical Method with Multipole Electrostatics and Density-Dependent Dispersion Contributions.J. Chem. Theory Comput.15, 1652–1671 (2019)

work page 2019
[36]

Bannwarth, C.et al.Extended tight-binding quantum chemistry methods.Wiley Interdiscip. Rev. Comput. Mol. Sci.11, 1–49 (2021)

work page 2021
[37]

& Grimme, S

Spicher, S. & Grimme, S. Robust Atomistic Modeling of Materials, Organometallic, and Biochemical Systems. Angew. Chem. Int. Ed.132, 15795–15803 (2020)

work page 2020
[38]

& Shushkov, P

Grimme, S., Bannwarth, C. & Shushkov, P. A Robust and Accurate Tight-Binding Quantum Chemical Method for Structures, Vibrational Frequencies, and Noncovalent Interactions of Large Molecular Systems Parametrized for All spd-Block Elements (Z = 1-86).J. Chem. Theory Comput.13, 1989–2009 (2017)

work page 1989
[39]

The PyMOL molecular graphics system, version 1.8 (2015)

Schrödinger, LLC. The PyMOL molecular graphics system, version 1.8 (2015)

work page 2015
[40]

Kim, S.et al.Pubchem 2025 update.Nucleic Acids Res.53, D1516–D1525 (2025)

work page 2025
[41]

& Gómez-Bombarelli, R

Axelrod, S. & Gómez-Bombarelli, R. GEOM, energy-annotated molecular conformations for property prediction and molecular generation.Sci. Data9(2022)

work page 2022
[42]

M.et al.Training a Scientific Reasoning Model for Chemistry (2025)

Narayanan, S. M.et al.Training a Scientific Reasoning Model for Chemistry (2025). arXiv:2506.17238 [cs]

work page arXiv 2025
[43]

InProceedings of the AAAI Conference on Artificial Intelligence, vol

Li, J.et al.Chemvlm: Exploring the power of multimodal large language models in chemistry area. InProceedings of the AAAI Conference on Artificial Intelligence, vol. 39, 415–423 (2025)

work page 2025
[44]

F.et al.Ucsf chimerax: Structure visualization for researchers, educators, and developers.Protein Sci.30, 70–82 (2021)

Pettersen, E. F.et al.Ucsf chimerax: Structure visualization for researchers, educators, and developers.Protein Sci.30, 70–82 (2021)

work page 2021
[45]

C.et al.UCSF ChimeraX: Tools for structure building and analysis.Protein Sci.32, e4792 (2023)

Meng, E. C.et al.UCSF ChimeraX: Tools for structure building and analysis.Protein Sci.32, e4792 (2023)

work page 2023
[46]

& Kondo, M

Imai, M., Masaoka, S. & Kondo, M. Function provision to fe–porphyrin complexes for CO2 reduction.JACS Au 5, 3716–3737 (2025)

work page 2025
[47]

Catal.7, 1000–1009 (2024)

Yu, S.et al.CO 2-to-methanol electroconversion on a molecular cobalt catalyst facilitated by acidic cations.Nat. Catal.7, 1000–1009 (2024)

work page 2024
[48]

Han, N.et al.Supported cobalt polyphthalocyanine for high-performance electrocatalytic CO2 reduction.Chem 3, 652–664 (2017)

work page 2017
[49]

Catal.7, 987–999 (2024)

Zhu, Q.et al.The solvation environment of molecularly dispersed cobalt phthalocyanine determines methanol selectivity during electrocatalytic CO2 reduction.Nat. Catal.7, 987–999 (2024)

work page 2024
[50]

Su, J.et al.Strain enhances the activity of molecular electrocatalysts via carbon nanotube supports.Nat. Catal. 6, 818–828 (2023). 29

work page 2023
[51]

Catal.6, 234–243 (2023)

Lee, B.-H.et al.Supramolecular tuning of supported metal phthalocyanine catalysts for hydrogen peroxide electrosynthesis.Nat. Catal.6, 234–243 (2023)

work page 2023
[52]

& McCrory, C

Liu, Y. & McCrory, C. C. Modulating the mechanism of electrocatalytic CO2 reduction by cobalt phthalocyanine through polymer coordination and encapsulation.Nat. Commun.10, 1683 (2019)

work page 2019
[53]

Commun.14, 1426 (2023)

Fan, W.et al.Rational design of heterogenized molecular phthalocyanine hybrid single-atom electrocatalyst towards two-electron oxygen reduction.Nat. Commun.14, 1426 (2023)

work page 2023
[54]

& Kulik, H

Nandy, A., Duan, C., Goffinet, C. & Kulik, H. J. New strategies for direct methane-to-methanol conversion from active learning exploration of 16 million catalysts.JACS Au2, 1200–1213 (2022)

work page 2022
[55]

Bursch, M.et al.Understanding and quantifying london dispersion effects in organometallic complexes.Acc. Chem. Res.52, 258–266 (2018)

work page 2018
[56]

& Nishibayashi, Y

Arashiba, K., Miyake, Y. & Nishibayashi, Y. A molybdenum complex bearing PNP-type pincer ligands leads to the catalytic reduction of dinitrogen into ammonia.Nat. Chem.3, 120–125 (2011)

work page 2011
[57]

Chem.16, 22–27 (2024)

Kuramochi, H.et al.Ultrafast Raman observation of the perpendicular intermediate phantom state of stilbene photoisomerization.Nat. Chem.16, 22–27 (2024)

work page 2024
[58]

& Wezenberg, S

Villarón, D. & Wezenberg, S. J. Stiff-stilbene photoswitches: from fundamental studies to emergent applications. Angew. Chem. Int. Ed.132, 13292–13302 (2020)

work page 2020
[59]

Software update: The ORCA program system—version 6.0.Wiley Interdiscip

Neese, F. Software update: The ORCA program system—version 6.0.Wiley Interdiscip. Rev. Comput. Mol. Sci. 15, e70019 (2025)

work page 2025
[60]

Max-Planck-Institut für Kohlenforschung (2025)

Neese, F.ORCA 6.0 Manual. Max-Planck-Institut für Kohlenforschung (2025). Version 6.0

work page 2025
[61]

Alt, H. G. & Köppl, A. Effect of the nature of metallocene complexes of group iv metals on their performance in catalytic ethylene and propylene polymerization.Chem. Rev.100, 1205–1222 (2000)

work page 2000
[62]

G., Froese, R

Musaev, D. G., Froese, R. D., Svensson, M. & Morokuma, K. A density functional study of the mechanism of the diimine- nickel-catalyzed ethylene polymerization reaction.J. Am. Chem. Soc.119, 367–374 (1997)

work page 1997
[63]

Becke, A. D. Density-functional thermochemistry. III. The role of exact exchange.J. Chem. Phys.98, 5648–5652 (1993)

work page 1993
[64]

& Parr, R

Lee, C., Yang, W. & Parr, R. G. Development of the Colle-Salvetti correlation-energy formula into a functional of the electron density.Physical Review B37, 785–789 (1988)

work page 1988
[65]

J., Devlin, F

Stephens, P. J., Devlin, F. J., Chabalowski, C. F. & Frisch, M. J. Ab Initio Calculation of Vibrational Absorption and Circular Dichroism Spectra Using Density Functional Force Fields.J. Phys. Chem.98, 11623–11627 (1994)

work page 1994
[66]

& Grimme, S

Caldeweyher, E., Bannwarth, C. & Grimme, S. Extension of the d3 dispersion coefficient model.J. Chem. Phys. 147(2017)

work page 2017
[67]

Caldeweyher, E.et al.A generally applicable atomic-charge dependent London dispersion correction.J. Chem. Phys.150, 154122 (2019)

work page 2019
[68]

& Ahlrichs, R

Weigend, F. & Ahlrichs, R. Balanced basis sets of split valence, triple zeta valence and quadruple zeta valence quality for H to Rn: Design and assessment of accuracy.Phys. Chem. Chem. Phys.7, 3297–3305 (2005)

work page 2005
[69]

Fanourakis, A.et al.Strategic atom replacement enables regiocontrol in pyrazole alkylation.Nature641, 646–652 (2025)

work page 2025
[70]

R., Bruno, I

Groom, C. R., Bruno, I. J., Lightfoot, M. P. & Ward, S. C. The cambridge structural database.Acta Crystallogr. Sect. B Struct. Sci. Cryst. Eng. Mater.72, 171–179 (2016)

work page 2016
[71]

S.et al.The open molecules 2025 (omol25) dataset, evaluations, and models (2025)

Levine, D. S.et al.The open molecules 2025 (omol25) dataset, evaluations, and models (2025). arXiv:2505.08762 [physics.chem-ph]

work page arXiv 2025
[72]

Guo, D.et al.DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning.Nature645, 633–638 (2025)

work page 2025
[73]

Ouyang, L.et al.Training language models to follow instructions with human feedback.Advances in neural information processing systems35, 27730–27744 (2022)

work page 2022
[74]

G.et al.Architector for high-throughput cross-periodic table 3D complex building.Nat

Taylor, M. G.et al.Architector for high-throughput cross-periodic table 3D complex building.Nat. Commun.14, 2786 (2023). 30

work page 2023
[75]

chemrxiv-2025-h8gff-v2

Terrones, G.et al.molSimplify 2.0: Improved structure generation for automating discovery in inorganic molecular and reticular chemistry (2025). chemrxiv-2025-h8gff-v2

work page 2025
[76]

Matter8(2025)

Darvish, K.et al.Organa: A robotic assistant for automated chemistry experimentation and characterization. Matter8(2025)

work page 2025
[77]

Raucci, U.et al.Voice-controlled quantum chemistry.Nat. Comput. Sci.1, 42–45 (2021)

work page 2021
[78]

& Reiher, M

Aspuru-Guzik, A., Lindh, R. & Reiher, M. The matter simulation (R)evolution.ACS Cent. Sci.4, 144–152 (2018)

work page 2018
[79]

V.et al.Optimising digital twin laboratories with conversational ais: enhancing immersive training and simulation through virtual reality.Digital Discovery4, 1134–1141 (2025)

Taylor, M. V.et al.Optimising digital twin laboratories with conversational ais: enhancing immersive training and simulation through virtual reality.Digital Discovery4, 1134–1141 (2025)

work page 2025
[80]

O’Connor, M.et al.Sampling molecular conformations and dynamics in a multiuser virtual reality framework. Sci. Adv.4, eaat2731 (2018)

work page 2018

Showing first 80 references.