RoboFail

The 'RoboFail' dataset/benchmark contains diverse robotic manipulation scenarios used to evaluate the effectiveness of reinforcement learning in enhancing process reasoning for long-horizon tasks.

Papers using RoboFail (2)

From Passive Observer to Active Critic: Reinforcement Learning Elicits Process Reasoning for Robotic Manipulation2026Scaling Cross-Environment Failure Reasoning Data for Vision-Language Robotic Manipulation2025