Qin and colleagues: ToolLLM data, training and tool-use evaluation

What changed

Constructed tool-use instructions and solution traces, trained a tool-oriented model with API retrieval, and evaluated multi-step API use.

Scientific scope

Constructed tool-use instructions and solution traces, trained a tool-oriented model with API retrieval, and evaluated multi-step API use. Synthetic teacher-generated paths and model-based evaluation introduce bias; API availability can change. ToolBench, ToolLLaMA and ToolEval are kept in one research package.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Synthetic teacher-generated paths and model-based evaluation introduce bias; API availability can change. ToolBench, ToolLLaMA and ToolEval are kept in one research package.

Contributors

• Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.

• ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction