_version_ 1866916463900622848
author Mirza, Adrian
Alampara, Nawaf
Kunchapu, Sreekanth
Ríos-García, Martiño
Emoekabu, Benedict
Krishnan, Aswanth
Gupta, Tanya
Schilling-Wilhelmi, Mara
Okereke, Macjonathan
Aneesh, Anagha
Elahi, Amir Mohammad
Asgari, Mehrdad
Eberhardt, Juliane
Elbeheiry, Hani M.
Gil, María Victoria
Greiner, Maximilian
Holick, Caroline T.
Glaubitz, Christina
Hoffmann, Tim
Ibrahim, Abdelrahman
Klepsch, Lea C.
Köster, Yannik
Kreth, Fabian Alexander
Meyer, Jakob
Miret, Santiago
Peschel, Jan Matthias
Ringleb, Michael
Roesner, Nicole
Schreiber, Johanna
Schubert, Ulrich S.
Stafast, Leanne M.
Wonanke, Dinga
Pieler, Michael
Schwaller, Philippe
Jablonka, Kevin Maik
author_facet Mirza, Adrian
Alampara, Nawaf
Kunchapu, Sreekanth
Ríos-García, Martiño
Emoekabu, Benedict
Krishnan, Aswanth
Gupta, Tanya
Schilling-Wilhelmi, Mara
Okereke, Macjonathan
Aneesh, Anagha
Elahi, Amir Mohammad
Asgari, Mehrdad
Eberhardt, Juliane
Elbeheiry, Hani M.
Gil, María Victoria
Greiner, Maximilian
Holick, Caroline T.
Glaubitz, Christina
Hoffmann, Tim
Ibrahim, Abdelrahman
Klepsch, Lea C.
Köster, Yannik
Kreth, Fabian Alexander
Meyer, Jakob
Miret, Santiago
Peschel, Jan Matthias
Ringleb, Michael
Roesner, Nicole
Schreiber, Johanna
Schubert, Ulrich S.
Stafast, Leanne M.
Wonanke, Dinga
Pieler, Michael
Schwaller, Philippe
Jablonka, Kevin Maik
contents Large language models (LLMs) have gained widespread interest due 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 outperformed the best human chemists in our study on average. 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.
format Preprint
id arxiv_https___arxiv_org_abs_2404_01475
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Are large language models superhuman chemists?
Mirza, Adrian
Alampara, Nawaf
Kunchapu, Sreekanth
Ríos-García, Martiño
Emoekabu, Benedict
Krishnan, Aswanth
Gupta, Tanya
Schilling-Wilhelmi, Mara
Okereke, Macjonathan
Aneesh, Anagha
Elahi, Amir Mohammad
Asgari, Mehrdad
Eberhardt, Juliane
Elbeheiry, Hani M.
Gil, María Victoria
Greiner, Maximilian
Holick, Caroline T.
Glaubitz, Christina
Hoffmann, Tim
Ibrahim, Abdelrahman
Klepsch, Lea C.
Köster, Yannik
Kreth, Fabian Alexander
Meyer, Jakob
Miret, Santiago
Peschel, Jan Matthias
Ringleb, Michael
Roesner, Nicole
Schreiber, Johanna
Schubert, Ulrich S.
Stafast, Leanne M.
Wonanke, Dinga
Pieler, Michael
Schwaller, Philippe
Jablonka, Kevin Maik
Machine Learning
Materials Science
Artificial Intelligence
Chemical Physics
Large language models (LLMs) have gained widespread interest due 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 outperformed the best human chemists in our study on average. 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.
title Are large language models superhuman chemists?
topic Machine Learning
Materials Science
Artificial Intelligence
Chemical Physics
url https://arxiv.org/abs/2404.01475