Inconsistency of LLMs in molecular representations
We examine whether LLMs capture the underlying chemistry beyond symbolic representations by evaluating task consistency when using SMILES strings vs. IUPAC names as inputs.
Authors
- Angelica Chen
- Bing Yan (ORCID: https://orcid.org/0000-0002-3489-7253)
- Kyunghyun Cho
Institutions
- New York University (US)
Publication Details
- Journal
- Digital Discovery
- Published
- 2025-01-01
- DOI
- https://doi.org/10.1039/d5dd00176e
- Citations
- 1
- Primary Topic
- Machine Learning in Materials Science
- Type
- article
- Field-Weighted Citation Impact
- 0.52
Funders
- Samsung Advanced Institute of Technology
- Ministry of Science and ICT, South Korea
- Samsung
- Institute for Information and Communications Technology Promotion