Paper recorded by Signals 4 on 2026-09-30 in cs.AI. Abstract reproduced from arXiv; link to the original below.
Published 2026-09-30 on arXiv · recorded by Signals 4 on 2026-10-01
Category: cs.AI · 人工智能 · first seen 2026-10-01
Online training enables computer-use agents (CUAs) to improve through interaction with executable environments. However, existing methods primarily rely on sparse outcome rewards, which provide no supervision for intermediate actions. On-policy self-distillation (OPSD) offers token-level learning signals through privileged rescoring, but directly applying it to CUA online training presents two cha