GOAT synthesizes goal-oriented API execution training data from API documents via a call-first generation paradigm, enabling fine-tuned agents to achieve state-of-the-art performance on benchmarks while introducing GOATBench.
- If a parameter cannot be filled this way, infer it using the information in the API Document (e.g., parameter descriptions or type hints)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
GOAT: A Training Framework for Goal-Oriented Agent with Tools
GOAT synthesizes goal-oriented API execution training data from API documents via a call-first generation paradigm, enabling fine-tuned agents to achieve state-of-the-art performance on benchmarks while introducing GOATBench.