STeCa: Step-level Trajectory Calibration for LLM Agent Learning.

Hanlin Wang, Jian Wang, Chak Tou Leong, Wenjie Li

Browse the full ACL paper archive.