arXiv · 机器人与计算机视觉· Dongsu Lee, Haoran Xu, Amy Zhang·· 2026-10-05AI 评分75
Test-time Multi-agent Coordination by Decomposed Value Gradient Flow
Test-time Multi-agent Coordination by Decomposed Value Gradient Flow
AI 导读
arXiv 论文《Test-time Multi-agent Coordination by Decomposed Value Gradient Flow》由 Dongsu Lee、Haoran Xu 和 Amy Zhang 提交至 NeurIPS 2026,提出 SCOUT 框架以解决离线多智能体强化学习(MARL)中生成策略与价值优化策略的权衡问题。
来源:arXiv · 机器人与计算机视觉 · arxiv.org