Paper recorded by Signals 4 on 2026-10-09 in cs.AI. Abstract reproduced from arXiv; link to the original below.
Published 2026-10-09 on arXiv · recorded by Signals 4 on 2026-10-12
Category: cs.AI · 人工智能 · first seen 2026-10-12
Training long-horizon language model agents requires assigning credit to intermediate decisions from sparse terminal feedback. Within a finite group of sibling rollouts, useful action comparisons depend on their context, evidence for both alternatives, and how that evidence enters learning. We propose Matched-Context Group Policy Optimization (MCGPO), a procedure for constructing weighted action p