Fine-tuning LLMs on a small set of instruction data synthesized from knowledge graphs substantially improves their tool-use performance on T-Eval.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Enhancing LLM Tool Use with High-quality Instruction Data from Knowledge Graph
Fine-tuning LLMs on a small set of instruction data synthesized from knowledge graphs substantially improves their tool-use performance on T-Eval.