记忆归属何处：Ledger——面向记忆增强型VLA的对象账本
Where Memory Belongs: Ledger, an Object Ledger for Memory-Augmented VLAs
- Harvard University（哈佛大学）
- ETH Zürich（苏黎世联邦理工学院）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对长时程机器人操作，提出Ledger框架，将短期感知记忆置于策略内、长期对象记忆置于外部账本，在RoboMME上以64.3%平均分显著超越先前方法。
AI中文摘要：
记忆对于长时程、部分可观测的机器人操作至关重要：机器人必须记住哪个物体被放入了抽屉、移动了谁的杯子，或者已经过去了多少个动作周期。近期的视觉-语言-动作（VLA）模型将记忆直接嵌入策略内部，但基准测试表明，没有任何单一的策略内机制能够覆盖所有时空维度，与理想方法相比差距悬殊。我们认为，记忆类型决定了记忆应存放的位置：短期感知记忆（重复、时序、回溯）应置于策略内部，而长期对象记忆（持久空间状态、包含关系、事件历史）则应置于策略外部，作为显式、可读的记录。我们提出了Ledger，一种通过将策略内帧采样记忆与外部时空对象记忆（即账本）配对，在单个微调后的$\pi_{0.5}$策略上实现这一划分的框架。账本由SAM3跟踪器和VLM描述器从演示中构建，并由LLM规划器在步骤边界读取并做出决策。在RoboMME上，Ledger在所评估方法中取得了最高的四套件平均分，达64.3%（而同等评估条件下最强先前方法为45.9%），在对象引用（60.7%对40.3%）和对象恒存（86.7%对56.2%）上均领先，且仅使用一组权重。在运行时根据指令和记录选择记忆来源，无需任务级路由器。
英文摘要：
Memory is essential for long-horizon, partially observed robotic manipulation: a robot must remember which object was placed in a drawer, whose cup it moved, or how many action cycles have elapsed. Recent vision-language-action (VLA) models embed memory directly inside the policy, but benchmarks show no single in-policy mechanism covers all spatio-temporal dimensions, trailing oracle methods by a wide margin. We argue that memory type dictates where memory should reside: short-term perceptual memory (repetition, timing, retracing) belongs inside the policy, while long-term object memory (persistent spatial state, containment, event history) belongs outside as an explicit, readable record. We present Ledger, a harness that realizes this split over a single fine-tuned $π_{0.5}$ policy by pairing an in-policy frame-sampling memory with an external spatio-temporal object memory, the ledger, built from a SAM3 tracker and a VLM captioner of the demonstration and read by an LLM planner that decides at step boundaries. On RoboMME, Ledger reaches the highest four-suite average among the evaluated methods, 64.3% (vs. 45.9% for the strongest prior method under identical evaluation), leading object reference (60.7% vs. 40.3%) and object permanence (86.7% vs. 56.2%) using a single set of weights. Choosing the memory source at runtime, from the instruction and the record, removes the need for a task-level router.
