Oriel — AI & technology worth noticing, every day.

Don't Just Impute, Detect: Task-Aware Diffusion Pipelines for Edge Anomaly Detection

Paper · https://arxiv.org/abs/2610.40732

arXiv:2610.40732v1 Announce Type: new Abstract: Industrial Internet of Things (IIoT) edge pipelines routinely lose contiguous blocks of multi-sensor data to network dropouts and hardware faults. The standard remedy is a disjointed "impute-then-detect" pipeline whose imputer is trained only for Mean Squared Error (MSE). Such an imputer fills a missing block with the most probable normal pattern: a partially observed anomaly is smoothed back toward normality, and boundary artifacts on normal data raise false alarms. We call this imputation bias and propose the Task-Aware Diffusion Refiner (TADR), a conditional diffusion refiner trained against the loss landscape of a frozen downstream detector. Because back-propagating through the sampling trajectory is intractable, the task-aware terms act on the one-step clean-window estimate that the noise prediction already yields, so training costs no more than plain diffusion training. Under strict Point-to-Point evaluation over five seeds with paired significance tests, TADR raises F1 by $0.338$ over the unrefined detector on SMD ($p = 0.0015$), and the anomaly enhancement objective alone contributes $0.463$ in ablation. We also characterise when the method helps: the gain tracks the continuous, mutually redundant channels available and the anomaly evidence the outage removes, and it is absent on telemetry datasets dominated by one-hot command channels and on PSM. At ten sampling steps TADR runs in $41.4$ ms per window on a single CPU core, sustaining about $24$ concurrent 1 Hz streams from a $4.03$ MB model. Code and per-seed results are available at https://github.com/int-park422/TADR.

First seen by Oriel on .

AI summary

The paper introduces Task-Aware Diffusion Refiner (TADR), which trains missing-data reconstruction for downstream anomaly detection and improves edge detection performance under contiguous IIoT outages.

  • TADR conditions a diffusion refiner on observed data and optimizes task-aware terms using a one-step clean-window estimate, avoiding back-propagation through sampling.
  • Across five seeds on SMD, TADR improves F1 by 0.338 over the unrefined detector, with paired significance testing yielding p = 0.0015.
  • An anomaly enhancement objective contributes 0.463 in ablation, while gains depend on redundant continuous channels and the anomaly evidence removed by outages.
  • The method provides no gain on telemetry dominated by one-hot command channels or on PSM, limiting its applicability across dataset types.
  • At ten sampling steps, TADR processes a window in 41.4 ms on one CPU core, supporting about 24 concurrent 1 Hz streams with a 4.03 MB model.

#diffusion models #anomaly detection #industrial iot #edge computing

Generated by gpt-5.6-sol. Check the original for details.

Aliases

SourceExternal IDURL
arXiv AI Papersoai:arXiv.org:2610.40732v1https://arxiv.org/abs/2610.40732 Read discussion