arXiv Machine Learning

A Real-Time Tsetlin Machine-based Non-intrusive Load Monitoring System on MCUs

The paper introduces a Tsetlin Machine (TM)-based framework for Non‑Intrusive Load Monitoring (NILM) that can run in real time on resource‑constrained microcontrollers. By reformulating NILM as a classification problem, the authors achieve high accuracy—90 % precision and 96 % recall for two appliances, and 77 % precision and 80 % recall for four appliances on the REDD dataset—while keeping the trained model to only 18 KB of flash memory. On an ESP32, the system delivers inference in 0.43 ms, demonstrating its feasibility for privacy‑preserving edge deployment.

arXiv Machine Learning
1d ago

Ask for Any Appliance: A Prompt-Programmable Foundation Model for Non-Intrusive Load Monitoring

The paper introduces FM4NILM, a prompt‑programmable foundation model that estimates the power usage of any requested appliance from whole‑home meter data, a natural‑language description, and optional activation examples. Trained on 645k sequences from seven public datasets, the lightweight transformer achieves competitive performance across twelve appliance requests, outperforming seven appliance‑specific baselines on key metrics such as event F1 and active‑window MAE. The model’s design allows new appliance coverage to be added via prompts and examples rather than building separate specialist networks.

By Xudong Wang, Jiacheng Cui, Junyu Xue, Tongxin Li, Guoming Tang
arXiv Machine Learning
6d ago

Multi-Appliance Non-Intrusive Load Monitoring via Label-Preserving Aggregate Recomposition and Prediction Consistency

The paper introduces a new NILM approach that uses label‑preserving aggregate recomposition and prediction consistency to improve appliance‑level power estimation. By recomposing aggregate windows with only the residual background changed, the method preserves target appliance signals while exposing a new supervisory signal. Experiments on REDD, UK‑DALE, and REFIT show reduced mean absolute error for multiple appliances without adding inference‑time complexity.

By Jiangfeng Liu, Yanfang Fan
arXiv Machine Learning
Jun 24

EnerInfer: Energy-Aware On-Device LLM Inference

arXiv:2606. 23001v1 Announce Type: cross Abstract: On-device LLM inference is increasingly attractive for privacy-preserving, reliable, and cost-effective deployment, yet its energy and thermal costs remain a critical bottleneck.

By Bohua Zou, Nian Liu, Binqi Sun, Matteo Mascherin, Debayan Roy, Yutao Liu, Yu Peng, Ning Jia, Haibo Chen
arXiv Machine Learning
Sep 11

LoaDiff: Conditional Generation of Electricity Consumption Time Series for Energy Analytics

LoaDiff is a diffusion-based generative model that produces year-long, sub-hourly smart‑meter electricity consumption time series. It can be conditioned on static household attributes like appliance ownership and dynamic factors such as calendar dates and outdoor temperature. Evaluations on three residential datasets show that LoaDiff generates realistic, diverse load profiles, limits memorization, retains useful information for downstream tasks, and responds coherently to conditioning changes.

By Mariia Baranova, Adrien Petralia, Etienne Le Naour, Nathan Etourneau, Guillaume Hofmann, Themis Palpanas
arXiv AI
Jun 16

The Energy Blind Spot: NVIDIA's Flagship Edge AI Hardware Cannot Support Process-Level Energy Attribution

arXiv:2605. 27599v2 Announce Type: replace-cross Abstract: Agentic AI workloads - where a single user goal triggers multi-step orchestration, tool calls, retries, and failure recovery - are being targeted for edge deployment, with NVIDIA, Dell, HP, ASUS, MSI, Acer, and Gigabyte all shipping GB10-based desktop AI systems in 2026.

By Deepak Panigrahy, Aakash Tyagi
arXiv Machine Learning
Sep 14

The Battery Price of edge AI: A study of the Environmental Impact of LLM Inference on Mobile Devices

The paper investigates the environmental impact of running large language models (LLMs) on mobile devices. It evaluates 18 different LLM configurations on two smartphones and a server, measuring energy per token, latency, accuracy, and battery-cycle consumption. Findings reveal that on-device inference is about three times less energy‑efficient than batched server inference, that energy consumption varies non‑monotonically with quantization bit‑width, and that most models are not on the Pareto front of accuracy and energy efficiency. The study concludes that local AI is not inherently more sustainable than cloud inference, with the majority of environmental impact stemming from device embodied carbon.

By \'Edouard Gu\'egain, Tristan Coignion