dorsal/arxiv
View SchemaWhen Smaller Wins: Dual-Stage Distillation and Pareto-Guided Compression of Liquid Neural Networks for Edge Battery Prognostics
| Authors | Dhivya Dharshini Kannan, Wei Li, Wei Zhang, Jianbiao Wang, Zhi Wei Seh, Man-Fai Ng |
|---|---|
| Categories | |
| ArXiv ID | 2601.06227vv2 |
| URL | https://arxiv.org/abs/2601.06227 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
Battery management systems increasingly require accurate battery health prognostics under strict on-device constraints. This paper presents DLNet, a practical framework with dual-stage distillation of liquid neural networks that turns a high-capacity model into compact and edge-deployable models for battery health prediction. DLNet first applies Euler discretization to reformulate liquid dynamics for embedded compatibility. It then performs dual-stage knowledge distillation to transfer the teacher model's temporal behavior and recover it after further compression. Pareto-guided selection under joint error-cost objectives retains student models that balance accuracy and efficiency. We evaluate DLNet on a widely used dataset and validate real-device feasibility on an Arduino Nano 33 BLE Sense using int8 deployment. The final deployed student achieves a low error of 0.0066 when predicting battery health over the next 100 cycles, which is 15.4% lower than the teacher model. It reduces the model size from 616 kB to 94 kB with 84.7% reduction and takes 21 ms per inference on the device. These results support a practical smaller wins observation that a small model can match or exceed a large teacher for edge-based prognostics with proper supervision and selection. Beyond batteries, the DLNet framework can extend to other industrial analytics tasks with strict hardware constraints.
{
"annotation_id": "a64c6b3a-02e0-48c9-85b9-56b1273c7fab",
"date_created": "2026-02-17T05:53:07.515000Z",
"date_modified": "2026-02-17T05:53:07.515000Z",
"file_hash": "77a57fa328d0bf2a732ff9017cfa5d6077e772cfda8a9020ecead87f695b2eac",
"private": false,
"record": {
"abstract": "Battery management systems increasingly require accurate battery health prognostics under strict on-device constraints. This paper presents DLNet, a practical framework with dual-stage distillation of liquid neural networks that turns a high-capacity model into compact and edge-deployable models for battery health prediction. DLNet first applies Euler discretization to reformulate liquid dynamics for embedded compatibility. It then performs dual-stage knowledge distillation to transfer the teacher model\u0027s temporal behavior and recover it after further compression. Pareto-guided selection under joint error-cost objectives retains student models that balance accuracy and efficiency. We evaluate DLNet on a widely used dataset and validate real-device feasibility on an Arduino Nano 33 BLE Sense using int8 deployment. The final deployed student achieves a low error of 0.0066 when predicting battery health over the next 100 cycles, which is 15.4% lower than the teacher model. It reduces the model size from 616 kB to 94 kB with 84.7% reduction and takes 21 ms per inference on the device. These results support a practical smaller wins observation that a small model can match or exceed a large teacher for edge-based prognostics with proper supervision and selection. Beyond batteries, the DLNet framework can extend to other industrial analytics tasks with strict hardware constraints.",
"arxiv_id": "2601.06227",
"authors": [
"Dhivya Dharshini Kannan",
"Wei Li",
"Wei Zhang",
"Jianbiao Wang",
"Zhi Wei Seh",
"Man-Fai Ng"
],
"categories": [
"cs.LG",
"cs.AI"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "When Smaller Wins: Dual-Stage Distillation and Pareto-Guided Compression of Liquid Neural Networks for Edge Battery Prognostics",
"url": "https://arxiv.org/abs/2601.06227",
"version": "v2"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "e0fd33c3-384d-4a35-b9eb-c7817b5f3fc5",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}