Martiño Ríos-García

Martiño Ríos-García

Auto-generated profile, based on publicly available information

Martiño Ríos-García

Laboratory of Organic and Macromolecular Chemistry (IOMC), Friedrich Schiller University Jena, Humboldtstraße 10, 07743 Jena, Germany

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 December 2024 in Chemical Society Reviews
Verified
Journal Article
Open access
PDF file
MSMara Schilling-Wilhelmi
Mara Schilling-Wilhelmi
MRMartiño Ríos-García
Martiño Ríos-García
SSSherjeel Shabih
Sherjeel Shabih
MGM. Victoria Gil
M. Victoria Gil
SMSantiago Miret
Santiago Miret
KJKevin Maik Jablonka
Kevin Maik Jablonka
Large language models (LLMs) allow for the extraction of structured data from unstructured sources, such as scientific papers, with unprecedented accuracy and performance.

20 May 2025 in Nature Chemistry
Verified
Journal Article
Open access
AMAdrian Mirza
Adrian Mirza
NANawaf Alampara
Nawaf Alampara
SKSreekanth Kunchapu
Sreekanth Kunchapu
MRMartiño Ríos-García
Martiño Ríos-García
BEBenedict Emoekabu
Benedict Emoekabu
KJKevin Maik Jablonka
Kevin Maik Jablonka
Large language models (LLMs) have gained widespread interest owing to their ability to process human language and perform tasks on which they have not been explicitly trained. However, we possess only a limited systematic understanding of the chemical capabilities of LLMs, which would be required to improve models and mitigate potential harm. Here we introduce ChemBench, an automated framework for evaluating the chemical knowledge and reasoning abilities of state-of-the-art LLMs against the expertise of chemists. We curated more than 2,700 question–answer pairs, evaluated leading open- and closed-source LLMs and found that the best models, on average, outperformed the best human chemists in our study. However, the models struggle with some basic tasks and provide overconfident predictions. These findings reveal LLMs’ impressive chemical capabilities while emphasizing the need for further research to improve their safety and usefulness. They also suggest adapting chemistry education and show the value of benchmarking frameworks for evaluating LLMs in specific domains.

11 August 2025 in Nature Computational Science
Verified
Journal Article
Open access
NANawaf Alampara
Nawaf Alampara
MSMara Schilling-Wilhelmi
Mara Schilling-Wilhelmi
MRMartiño Ríos-García
Martiño Ríos-García
IMIndrajeet Mandal
Indrajeet Mandal
PKPranav Khetarpal
Pranav Khetarpal
KJKevin Maik Jablonka
Kevin Maik Jablonka
Recent advancements in artificial intelligence have sparked interest in scientific assistants that could support researchers across the full spectrum of scientific workflows, from literature review to experimental design and data analysis. A key capability for such systems is the ability to process and reason about scientific information in both visual and textual forms—from interpreting spectroscopic data to understanding laboratory set-ups. Here we introduce MaCBench, a comprehensive benchmark for evaluating how vision language models handle real-world chemistry and materials science tasks across three core aspects: data extraction, experimental execution and results interpretation. Through a systematic evaluation of leading models, we find that although these systems show promising capabilities in basic perception tasks—achieving near-perfect performance in equipment identification and standardized data extraction—they exhibit fundamental limitations in spatial reasoning, cross-modal information synthesis and multi-step logical inference. Our insights have implications beyond chemistry and materials science, suggesting that developing reliable multimodal AI scientific assistants may require advances in curating suitable training data and approaches to training those models.

5 February 2026 in Chemical Reviews
Verified
Journal Article
Open access
NANawaf Alampara
Nawaf Alampara
AAAnagha Aneesh
Anagha Aneesh
MRMartiño Ríos-García
Martiño Ríos-García
AMAdrian Mirza
Adrian Mirza
MSMara Schilling-Wilhelmi
Mara Schilling-Wilhelmi
KJKevin Maik Jablonka
Kevin Maik Jablonka
Data-driven techniques have a large potential to transform and accelerate the chemical sciences. However, chemical sciences also pose the unique challenge of very diverse, small, fuzzy data sets that are difficult to leverage in conventional machine learning approaches. A new class of models, which can be summarized under the term general-purpose models (GPMs) such as large language models, has shown the ability to solve tasks they have not been directly trained on, and to flexibly operate with low amounts of data in different formats. In this review, we discuss the fundamental building principles of GPMs and review recent and emerging applications of those models in the chemical sciences. While many of these applications are still in the prototype phase, we expect that the increasing interest in GPMs will make many of them mature in the coming years.

3 March 2022 in Molecules
Verified
Journal Article
Open access
PDF file
MRMartiño Ríos-García
Martiño Ríos-García
BFBerta Fernández
Berta Fernández
JRJesús Rodríguez-Otero
Jesús Rodríguez-Otero
ECEnrique M. Cabaleiro-Lago
Enrique M. Cabaleiro-Lago
SVSaulo Vazquez
Saulo Vazquez
Recently, we reported a new approach to develop pairwise analytical corrections to improve the description of noncovalent interactions, by approximate methods of electronic structures, such as semiempirical quantum mechanical (SQM) methods. In particular, and as a proof of concept, we used the PM6 Hamiltonian and we named the method PM6-FGC, where the FGC acronym, corresponding to Functional Group Corrections, emphasizes the idea that the corrections work for specific functional groups rather than for individual atom pairs. The analytical corrections were derived from fits to B3LYP-D3/def2-TZVP (reference). PM6 interaction energy differences, evaluated for a reduced set of small bimolecular complexes, were chosen as representatives of saturated hydrocarbons, carboxylic, amine and, tentatively, amide functional groups. For the validation, the method was applied to several complexes of well-known databases, as well as to complexes of diglycine and dialanine, assuming the transferability of amine group corrections to amide groups. The PM6-FGC method showed great potential but revealed significant inaccuracies for the description of some interactions involving the –NH2 group in amines and amides, caused by the inadequate selection of the model compound used to represent these functional groups (an NH3 molecule). In this work, methylamine and acetamide are used as representatives of amine and amide groups, respectively. This new selection leads to significant improvements in the calculation of noncovalent interactions in the validation set.

21 August 2025 in Nature Computational Science
Verified
Journal Article
Open access
PDF file
NANawaf Alampara
Nawaf Alampara
MSMara Schilling-Wilhelmi
Mara Schilling-Wilhelmi
MRMartiño Ríos-García
Martiño Ríos-García
IMIndrajeet Mandal
Indrajeet Mandal
PKPranav Khetarpal
Pranav Khetarpal
KJKevin Maik Jablonka
Kevin Maik Jablonka