ToolMisuseBench: A deterministic benchmark for tool-augmented Agents(huggingface.co)
huggingface.co
ToolMisuseBench: A deterministic benchmark for tool-augmented Agents
https://huggingface.co/datasets/sigdelakshey/ToolMisuseBench
https://huggingface.co/datasets/sigdelakshey/ToolMisuseBench
This dataset is intended for reproducible evaluation of agent tool-use behavior, not for training a general-purpose language model.