Article

AGT-GSPO trains embodied VLMs through verifiable interaction to achieve execution error recovery

The paper proposes the ERR-Env verifiable interaction environment, the ERR-Bench benchmark, and an online reinforcement learning method, Action-Grouped Turn-GSPO (AGT-GSPO), to train the 2B embodied VLM ERR-VLM to learn replanning and recovery after execution errors.

Part of the event: AGT-GSPO trains embodied VLMs to achieve error-recovery replanning

This page provides a summary and source link only.