MoleCode is a training-free, LLM-native representation that makes molecular graphs with explicit atoms, bonds, and topology directly readable and editable in language models, improving structural tasks over implicit string encodings.
Improving chemical understanding of llms via smiles parsing
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
q-bio.BM 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
MoleCode unlocks structural intelligence in large language models
MoleCode is a training-free, LLM-native representation that makes molecular graphs with explicit atoms, bonds, and topology directly readable and editable in language models, improving structural tasks over implicit string encodings.