发表机构
Michigan State University; University of California Los Angeles（密歇根州立大学; 加利福尼亚大学洛杉矶分校）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
Meta-Ctrl是一种约束解码框架，通过引入元标记分离句法与语义约束，在保证规划满足约束的同时保留语言模型的规划质量，在WAH-NL数据集和真实桌面机器人上均取得优于GPT-4的效果。
AI 中文摘要
大型语言模型（LLMs）可为机器人生成流畅的规划，但常违反执行所需满足的句法与语义约束；现有补救措施需在形式保证与规划质量间权衡：软方法（可供性评分、接地解码）无保证，而符号规划器（LLM+P）会丢弃语言模型（LM）的常识。本文提出Meta-Ctrl，一种约束解码框架，可保证编码的约束同时保留基础LM的规划质量。Meta-Ctrl引入元标记（meta-tokens）——接地动作的紧凑词汇表，在标记级强制执行句法约束，在动作级强制执行语义约束（前置条件、目标、顺序），这种精确分解将约束解码的内存需求从超107TB降至2GB以下。借助该框架，小型开放权重语言模型在原本排名垫底的场景中变得有竞争力：在WAH-NL数据集的LoTa-Bench协议下，它达到了报告的最高子目标成功率，超过了GPT-4的表现，且在具身智能体接口（Embodied Agent Interface）上实现了一致的性能提升。我们还在真实桌面机器人上验证了该框架，所有生成的规划从结构上满足其前置条件与目标。项目网站：this https URL
英文摘要
LLMs generate fluent plans for robots but routinely violate the syntactic and se8mantic constraints they must satisfy to execute, and existing remedies trade formal guarantees against plan quality: soft methods (affordance scoring, grounded decoding) give no guarantee, while symbolic planners (LLM+P) discard the LM's commonsense. We propose \textbf{Meta-Ctrl}, a constrained-decoding framework that guarantees the encoded constraints while preserving the base LM's plan quality. Meta-Ctrl introduces \emph{meta-tokens}---a compact vocabulary of grounded actions---enforcing syntax at the token level and semantics (preconditions, goals, ordering) at the action level, an exact factorization that cuts the memory of constrained decoding from over 107TB to under 2GB. With it, a small open-weight LM becomes competitive where it otherwise sits at the bottom of the leaderboard: on WAH-NL under the LoTa-Bench protocol it reaches the highest reported subgoal success rate, exceeding GPT-4's, with consistent gains across the Embodied Agent Interface. We further demonstrate it on a real tabletop robot, where every generated plan satisfies its preconditions and goals by construction. Project website: https://metactrlg.github.io
