A self-updating library of decomposed chemistry sub-tasks improves LLM accuracy on SciBench chemistry problems, with reported gains of up to 46 percentage points over direct reasoning.
Only conclude the used knowledge and formulas used in this whole task, do NOT contain numerical calculation process and results
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
ChemAgent: Self-updating Library in Large Language Models Improves Chemical Reasoning
A self-updating library of decomposed chemistry sub-tasks improves LLM accuracy on SciBench chemistry problems, with reported gains of up to 46 percentage points over direct reasoning.