TAU-bench
Emerging5papers using it
2025first seen
'Tau-Bench' is an airline benchmark dataset used to evaluate the performance of tool-calling agents in realistic customer service tasks.
Papers using TAU-bench (5)
- Self-Challenging Language Model AgentsMulti-Turn Reinforcement Learning for Tool-Calling Agents with Iterative Reward CalibrationCM2: Reinforcement Learning with Checklist Rewards for Multi-Turn and Multi-Step Agentic Tool UseRobust Tool Use via Fission-GRPO: Learning to Recover from Execution ErrorsGeneralizable End-to-End Tool-Use RL with Synthetic CodeGym