LLM agents autonomously generate, constrain, and simulate-test MOF design hypotheses across six tasks, concentrating on top structures within 400 evaluations while producing de novo candidates that beat random search and genetic algorithms.
The Journal of Physical Chemistry B 102, 2569–2577 (1998)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Interpretable Inverse Design of Metal-Organic Frameworks with Large Language Model Agents
LLM agents autonomously generate, constrain, and simulate-test MOF design hypotheses across six tasks, concentrating on top structures within 400 evaluations while producing de novo candidates that beat random search and genetic algorithms.