Get Started
Home
Topics
Search
Library
Research questionHow can retrieval-augmented factory assistants retain adapted answer quality across heterogeneous edge devices under memory, throughput, and capability constraints?Factory workers need conversational access to machine manuals, but capable models may not fit shop-floor hardware. Compression can affect general capability differently from retrieval-grounded answer quality, making device-specific deployment choices difficult.
AI
Inference Optimization
Natural Language Processing
Retrieval-Augmented Generation
Small / On-device Models
Latest papersRecent research connected to this question, newest first.Measurement-Driven Sub-Network Selection for On-Premise Retrieval-Augmented Factory AgentsThe evidence covers retrieval-grounded adaptation for manufacturing manuals, structural compression, and selection across three heterogeneous edge tiers. The reported case uses a weight-shared supernetwork with in-place distillation and measures judged answer quality, on-device throughput, memory constraints, and standby power; it does not establish performance beyond this setting.research paper · Sep 2, 2026
Related questions
How can robotic agents maintain queryable spatial-semantic memory for precise, low-latency answers in complex environments?How can long-horizon LLM agents preserve answer quality under tight prompt-token budgets?How can edge analog in-memory inference reduce energy while certifying errors from heterogeneous, imperfect accelerators?How can search agents learn when retrieval is necessary and ground answers in evidence without costly supervision?