arXiv cs.CL· Dmytro Kurdydyk, Aadi Palnitkar, Sebastian Scherer, Seungchan Kim·· 11 小时前AI 评分31
SPARK：通过物理重观测实现自我纠正的具身记忆框架
SPARK: Self-Correcting Embodied Memory via Physical Re-Observation
AI 导读
SPARK 是一个面向具身 RAG 的纠正性多回合框架，通过语义批评器标记知识库问题区域、定向规划器生成重访、选择性合并增删替换知识库信息。在四个 AirSim 户外地图和 158 道具身问答题上，单次纠正回合在两种退化视角条件下均提升 EQA 表现，存储对象比仅追加更新少 13-41%，单地图条件对最高提升 2.7 倍。纠正延伸至第 4 回合后，所有地图的 EQA 表现均超过其未退化参考水平。
正文 · 原文
Abstract:Retrieval-augmented generation (RAG) systems have seen growing adoption in embodied settings as language-queryable memories of the environment. However, existing approaches often assume a reliable single-episode knowledge base (KB) or accumulate observations across episodes without correcting prior errors. Such errors can persist through incorrect semantic labels, inaccurate object positions, and inconsistent spatial relationships. To address this, we present SPARK, a corrective multi-episode framework for embodied RAG. After the first episode, a semantic critic flags regions of the KB, a targeted planner generates revisits, and a selective merge adds, removes, or replaces KB information based on the new observations. We evaluate SPARK on four outdoor AirSim maps under two degraded viewpoint conditions, using perception-built KBs, a visibility-constrained oracle used only for evaluation, and a fixed set of 158 embodied question answering (EQA) questions scored before and after correction. A single corrective episode improves downstream EQA on every map under both conditions while storing 13-41% fewer objects than append-only updating, with gains up to 2.7x on one map-condition pair. Extending correction through Episode 4 brings every map above the EQA performance of its undegraded reference. Ablations show that the critic's three signals each contribute to performance, that targeted revisits achieve a larger mean gain than untargeted ones under the same budget, and that selective merging achieves gains comparable to append-only updating with a smaller KB. Overall, SPARK demonstrates the value of correcting embodied memory through targeted re-observation, particularly as embodied systems scale and the risk of accumulated errors increases.
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG) |
| Cite as: | arXiv:2611.00344 [cs.RO] |
| (or arXiv:2611.00344v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2611.00344 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Dmytro Kurdydyk [view email]
[v1]
Fri, 30 Oct 2026 20:24:55 UTC (15,612 KB)
来源：arXiv cs.CL · arxiv.org