BFCLv-4

The 'BFCLv-4' dataset/benchmark is a collection of high-quality, multi-turn function calling data used to evaluate the performance of large language models and autonomous agents in real-world tool interaction scenarios.

Papers using BFCLv-4 (3)

CurateEvo: Data-Curation Evolving for Agentic Post-Training2026SeeUPO: Sequence-Level Agentic-RL with Convergence Guarantees2026FunReason-MT Technical Report: Overcoming the Complexity Barrier in Multi-Turn Function Calling2025