mlsanityData to 14 Oct 2026

ARID: A Deployable Edge AI System for Structured Information Extraction from Industrial Maintenance Work Orders

arXiv 2609.23582 ↗First posted 20 Sep 2026cs.CL cs.AI cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Maintenance work orders must often be processed offline on embedded hardware, yet downstream software requires predictable structured output. We present ARID (Aviation-inspired Routing for Industrial Deployment), which extracts component, failure mode, symptom, and maintenance action into fixed-schema JSON on an 8 GB NVIDIA Jetson Orin NX. ARID combines conservative dual-teacher filtering, targeted noise-aware synthesis, one routing decision per work order, 4-bit inference, and grammar-constrained decoding. From 2,326 unlabeled OMIn records, it retains 716 training pairs and adds 99 topology-constrained records targeting action extraction. On 300 human-labeled records, ARID reaches 84.8% token-F1 on the reference stack and 82.9% on the deployed Jetson. Resident serving achieves 5,310/5,656 ms P50/P99 at 12.5 W. On zero-shot MaintNet transfer, semantic F1 falls to 46.4% while parser success remains at least 99.8%, showing that output validity transfers but field semantics do not.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 14 of its 20 references that are in the map. First posted on arXiv 20 Sep 2026.

Builds on

14 of 20 references.

References · 14

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  2. 2023Efficient Guided Generation for Large Language Modelscited by 200
  3. 2024Mixture of LoRA Expertscited by 181
  4. 2023EdgeMoE: Empowering Sparse Large Language Models on Mobile Devicescited by 69
  5. 2020NAT: Noise-Aware Training for Robust Neural Sequence Labelingcited by 17
  6. 2025CLONE: Customizing LLMs for Efficient Latency-Aware Inference at the Edgecited by 16
  7. 2025LD-MoLE: Learnable Dynamic Routing for Mixture of LoRA Expertscited by 13
  8. 2025EdgeLoRA: An Efficient Multi-Tenant LLM Serving System on Edge Devicescited by 11
  9. 2020MaintNet: A Collaborative Open-Source Library for Predictive Maintenance Language Resourcescited by 8
  10. 2026Which Quantization Should I Use? A Unified Evaluation of llama.cpp Quantization on Llama-3.1-8B-Instructcited by 8
  11. 2025Understanding the Performance and Power of LLM Inferencing on Edge Acceleratorscited by 7
  12. 2026MoLoRA: Composable Specialization via Per-Token Adapter Routingcited by 3
2 more
  1. 2025Trusted Knowledge Extraction for Operations and Maintenance Intelligencecited by 2
  2. 2025KEO: Knowledge Extraction on OMIn via Knowledge Graphs and RAG for Safety-Critical Aviation Maintenancecited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026EdgeFM: Efficient Edge Inference for Vision-Language Models
  2. 2026Needle in the Repo: A Benchmark for Maintainability in AI-Generated Repository Edits
  3. 2026PinSieve: Production Selective VLM Serving and a Governed Memory Flywheel for Enterprise Content-Quality Triage
  4. 2026A Hybrid Vision-Language Architecture for Automated Defect Reasoning and Report Generation in Industrial Inspection
  5. 2026Plant Memory: Technical Language Processing and a Verified Maintenance Ontology for Agent-Accessible Failure Knowledge from SAP PM Work Orders
  6. 2026OR-Space: A Full-Lifecycle Workspace Benchmark for Industrial Optimization Agents
  7. 2026Beyond Fluent Generation: A CPU Reliability Benchmark for MCP-Style Tool Calling in Sub-2B Small Language Models for Edge Deployment
  8. 2026TSCG: Deterministic Tool-Schema Compilation for Agentic LLM Deployments