Tool use / function calling
ToolSafe: Enhancing Tool Invocation Safety of LLM-based agents via Proactive Step-level Guardrail and Feedback
Current frontier — recent, not yet superseded in the knowledge base
0 papers critique it · 0 beat it on benchmarks
Recent methods in the same sub-problem, not yet superseded in the knowledge base — arXiv benchmark leaders, not vetted production recommendations.