AtomTree：基于 LLM 的分层具身指令跟随框架
AtomTree: A Hierarchical Framework for State-Aware Embodied Instruction Following with LLMs
一个叫 AtomTree 的新框架，让 LLM 干家务类具身任务时不再一口气定死计划，卡住了就回退重规划，ALFRED 上零样本跑赢了之前的 LLM 规划器。
AtomTree 是一个面向具身任务规划的分层框架，让 LLM 根据当前符号状态逐步决定是继续分解子目标还是输出底层动作，而不是一次性生成整份计划。框架内置轻量记忆模块总结过往执行结果供后续决策复用。当底层动作停滞时，AtomTree 会从父节点重新生成子树计划，避免沿错误轨迹持续漂移。在 ALFRED 基准的零样本评测中，AtomTree 超过了此前的 LLM 规划方法。
AtomTree: A Hierarchical Framework for State-Aware Embodied Instruction Following with LLMs
Embodied instruction following in interactive environments requires agents to interpret natural language commands and complete long-horizon tasks. Existing planners either rely on large labeled trajectories, or use few-shot and zero-shot LLM prompting that produces a whole plan in a single serial reasoning process. These approaches depend on linguistic common sense, commit to a plan from the initial instruction, and only update after failures, which weakens robustness and state awareness. We present AtomTree, a hierarchical framework for state-aware task planning with LLMs. AtomTree incrementally builds a tree of subgoals and actions, where the LLM decides at each step, from the current symbolic state, whether to further decompose or to output a grounded low-level action. A lightweight memory module summarizes past outcomes for reuse in later decisions. When a low-level action stalls, AtomTree triggers hierarchical backtracking replanning to return to the parent node and regenerate the subtree plan from the latest state, preventing continued drift along an incorrect trajectory. In zero-shot evaluation on the ALFRED benchmark, AtomTree outperforms prior LLM-based planners. Project page: https://zhaihaotian.github.io/atomtree-website/.