Current AI-based vision-language models can perceive content very well, but they reach their limits in more complex scientific processes. This is shown in a new study by Dr. Kevin Jablonka and his team, conducted with international partners and published in the journal ‘Nature Computational Science’. In this work, the researchers systematically examined for the first time how well modern AI models can process visual and textual information in chemistry and materials science.
Original publication: Alampara et al.: „Probing the limitations of multimodal language models for chemistry and materials research“, Nature Computational Science (2025), DOI: 10.1038/s43588-025-00836-3
For more information, please also view the press release from FSU Jena.