mlsanityData to 14 Oct 2026

SPINE: A Fault Injection Profiler for Quantized Neural Networks under Accumulated Faults

arXiv 2606.19526 ↗First posted 17 Jun 2026cs.ARcited by 0

Not on the map. Why?

Abstract (arXiv)

Deploying deep neural networks at the edge demands efficient inference under strict cost and power constraints. Quantized neural networks address these demands by replacing floating-point parameters with low-precision integers, yet their weights remain continuously exposed to radiation-induced bit-flips during inference. Fault Injection can be used to simulate those environments, but existing studies fail to characterize how accumulated upsets translate into mispredictions under realistic memory layouts. This paper presents a GDB-driven profiling framework that injects cumulative weight bit-flips directly onto the target binary of edge CPUs, generating per-layer fault profiles without requiring model retraining or code modification. Evaluated across multiple topologies, quantization efforts, and memory layouts, the results indicate how selective hardening strategies should be applied to effectively protect neural networks.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 4 of its 24 references that are in the map. First posted on arXiv 17 Jun 2026.

Builds on

4 of 24 references.

References · 4

  1. 2016Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activationscited by 426
  2. 2015DeepSat - A Learning framework for Satellite Imagerycited by 38
  3. 2019When Single Event Upset Meets Deep Neural Networks: Observations, Explorations, and Remediescited by 15
  4. 2023NeuralFuse: Learning to Recover the Accuracy of Access-Limited Neural Network Inference in Low-Voltage Regimescited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2023FAQ: Mitigating the Impact of Faults in the Weight Memory of DNN Accelerators through Fault-Aware Quantization
  2. 2022FlatENN: Train Flat for Enhanced Fault Tolerance of Quantized Deep Neural Networks
  3. 2024Memory Faults in Activation-sparse Quantized Deep Neural Networks: Analysis and Mitigation using Sharpness-aware Training
  4. 2024ZOBNN: Zero-Overhead Dependable Design of Binary Neural Networks with Deliberately Quantized Parameters
  5. 2026From Bit-Position Sensitivity to Unequal Error Protection for DNN Inference Memory
  6. 2026RESQ: A Unified Framework for REliability- and Security Enhancement of Quantized Deep Neural Networks
  7. 2024DeepNcode: Encoding-Based Protection against Bit-Flip Attacks on Neural Networks
  8. 2024Analyzing the Single Event Upset Vulnerability of Binarized Neural Networks on SRAM FPGAs