ReTerm: Recovery Training for Self-Improving Terminal Agents
Researchers propose ReTerm, a self-improving terminal agent that can recover from errors during long-running execution. ReTerm is trained to learn from its own failed tasks and uses a customized data pipeline and online RL algorithm to improve its recovery capabilities. The model achieves a 14.6 point improvement over the base model on Terminal-Bench 2.0.
Save an API key to vote.