以奖励为中心的ReST-MCTS：高不确定性环境下机器人操作的鲁棒决策框架
Reward-Centered ReST-MCTS: A Robust Decision-Making Framework for Robotic Manipulation in High Uncertainty Environments
AI总结：
提出Reward-Centered ReST-MCTS框架，通过分解中间反馈为中心信号并引导搜索，解决高不确定性环境下MCTS因稀疏奖励和噪声导致的决策脆弱性问题，实验验证其在同骨干VLA上的鲁棒性。
摘要：
蒙特卡洛树搜索因能通过模拟改进动作选择而无需完全可微策略，在机器人操作中具有吸引力。然而，在不确定领域中，稀疏的终端奖励和噪声转换可能使浅层搜索变得脆弱：许多候选分支在后期展开前仍难以区分，而小的模拟预算放大了这种模糊性。本文提出以奖励为中心的ReST-MCTS，一种将中间反馈分解为规则、启发式、可选神经和价值估计通道的决策框架，将所得过程信号与匹配的任务上下文对齐，并利用它来偏置或修复搜索，同时保留终端任务评估。主要证据有意分层：局部任务和匹配的ManiSkill诊断隔离奖励中心机制和消融；匹配的选项级ManiSkill扫描测试在原始失败、观测噪声和初始位姿偏移下的鲁棒性，但不声称标准基准优越性；官方同骨干OpenVLA-OFT/LIBERO桥测试有界VLA动作修复。OpenVLA-OFT干净复现无论是否使用RCRM-Guard均达到10/10 LIBERO-Spatial成功率。单个套件同骨干动作通道压力伪影在十次配对LIBERO-Spatial动作通道压力情节中记录0/10未防护成功和9/10防护成功。额外的观测噪声、语言扰动和视觉干扰探测作为覆盖和负结果上下文报告，而非优越性证据。最终声明有界：Reward-Centered ReST-MCTS是同骨干高不确定性操作的可检查测试时验证器，而非替代VLA策略或广泛标准基准优越性声明。
英文摘要：
Monte Carlo tree search is attractive for robotic manipulation because it can improve action selection through simulation without requiring a fully differentiable policy. In uncertain domains, however, sparse terminal rewards and noisy transitions can make shallow search brittle: many candidate branches remain indistinguishable until late rollouts, and small simulation budgets amplify this ambiguity. This paper presents Reward-Centered ReST-MCTS, a decision-making framework that decomposes intermediate feedback into rule, heuristic, optional neural, and value-estimation channels, centers the resulting process signal against matched task contexts, and uses it to bias or repair search while preserving terminal-task evaluation. The primary evidence is intentionally tiered. Local tasks and matched ManiSkill diagnostics isolate reward-center mechanisms and ablations; matched option-level ManiSkill sweeps test robustness under primitive failure, observation noise, and initial-pose shifts while not claiming standard benchmark superiority; and an official same-backbone OpenVLA-OFT/LIBERO bridge tests bounded VLA action repair. The OpenVLA-OFT clean reproduction reaches 10/10 LIBERO-Spatial successes both with and without RCRM-Guard. A single-suite same-backbone action-channel stress artifact over ten paired LIBERO-Spatial action-channel stress episodes records 0/10 unguarded successes and 9/10 guarded successes. Additional observation-noise, language-perturbation, and visual-distractor probes are reported as coverage and negative-result context rather than superiority evidence. The resulting claim is bounded: Reward-Centered ReST-MCTS is an inspectable test-time verifier for same-backbone high-uncertainty manipulation, not a replacement VLA policy or a broad standard-benchmark superiority claim.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
