AI RESEARCH

Beyond the Proxy: Trajectory-Distilled Guidance for Offline GFlowNet Training

arXiv CS.AI

ArXi:2505.20110v3 Announce Type: replace-cross Generative Flow Networks (GFlowNets) excel at sampling diverse, high-reward objects. In many practical applications where active reward queries are infeasible, these models must be trained using static offline datasets. Prevailing