BFCL v-3

The 'BFCL v3' dataset/benchmark contains synthetic tool-use trajectories designed to evaluate the performance of reinforcement learning algorithms in executing tool-calling policies within complex environments.

Papers using BFCL v-3 (5)

Controllable and Verifiable Tool-Use Data Synthesis for Agentic Reinforcement Learning2026ToolSample: Dual Dynamic Sampling Methods with Curriculum Learning for RL-based Tool Learning2025ShiQ: Bringing back Bellman to LLMs2025MagicAgent: Towards Generalized Agent Planning2026TopoCurate:Modeling Interaction Topology for Tool-Use Agent Training2026