Web-Shepherd: Advancing PRMs for Reinforcing Web Agents
Hyungjoo Chae, Sunghwan Kim, Junhee Cho, Seungone Kim, Seungjun Moon, Gyeom Hwangbo, Dongha Lim, Minjin Kim, Yeonjun Hwang, Minju Gwak, Dongwook Choi, Minseok Kang, Gwanhoon Im, ByeongUng Cho, Hyojun Kim, Jun Hee Han, Taeyoon Kwon, Minju Kim, Beong-woo Kwak, Dongjin Kang, Jinyoung Yeo
arXiv·2025
WEB-SHEPHERD introduces the first process reward model (PRM) for web navigation, providing fine-grained, step-level feedback to web agents through a novel checklist-based approach. This method...