NVIDIA Industrial ReMEmbR：语义交通图71.3%，机器人车队共享时空记忆系统
【US】车队记忆别写时全字幕：NVIDIA Industrial ReMEmbR，语义交通图 71.3%
反直觉：仓机车队要复用语义记忆，写时全量字幕会爆算力/带宽；应按需解码。现有错处：中心化全字幕入库。作者改动：边端编码检索嵌入/视觉 token/重建潜变量，服务端按需解码；PII 置零。硬数字：缓存 token 解码 **450/450** 复现未拆分模型输出；NaVQA 不可解析答 **4.5%→1.9%**（半参数量 agent）；语义交通图平衡准确率 **71.3%**。可迁移判断：**明川：车队/多机记忆用「写时嵌入、读时解码」架构，禁中心全字幕。不要并进 #16 溯源超图，也不要并进 #17 PerMIT。**
NVIDIA Industrial ReMEmbR (arXiv:2610.28624)
“View PDF HTML (experimental) Abstract: Warehouse robot fleets continuously observe operational conditions, but preserving these observations”
Abstract:Warehouse robot fleets continuously observe operational conditions, but preserving these observations as reusable semantic memory incurs substantial computation, communication, and storage costs. For example, explaining recurring aisle blockages can require reviewing video from multiple robots across several shifts. We present IndustrialRemembr, a shared spatiotemporal memory system that extends NVIDIA ReMEmbR to robot fleets and reduces server-side computation by generating scene descriptions only on demand. On each robot, an NVIDIA Jetson encodes video clips into retrieval embeddings with Cosmos-Embed1, vision tokens with Cosmos3-Edge, and reconstruction latents with the Wan2.2 VAE, without uploading raw camera frames. The encoding models are re-used to detect people in image patches, zeroing them out to support PII abstraction for GDPR-compliant deployments. We split Cosmos3-Edge at its vision-language boundary, placing the vision encoder on the robot and the language decoder on an on-premises server. A Nemotron agent retrieves memories by content, position, and time and queries the decoder using cached vision tokens. This design eliminates caption generation at write time, while cached-token decoding reproduces the unsplit model's output in all 450 tested decodes.
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI); Software Engineering (cs.SE) |
| Cite as: | arXiv:2610.28624 [cs.RO] |
| (or arXiv:2610.28624v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.28624 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jad Yammine [view email]
[v1]
Tue, 20 Oct 2026 17:58:32 UTC (6,542 KB)
来源：Agent Control Flash News · arxiv.org
关联产品与深度拆解
Cursor Agent 架构全景拆解：从影子工作区到多文件推演引擎
全面剖析 Cursor 在 Agent 模式下的底层技术实现，解析影子分支隔离、LSP 上下文剪枝以及预测性补全的核心架构细节与工程实践经验。
Vercel v0 推出全新多 Agent 协作画布与实时全栈部署预览
动态速评：v0 正在从单一组件生成器演进为全栈应用搭建协作台，尤其是与 Neon/Postgres 的直连极大降低了全栈原型验证成本。