TreeFI is a value‑aware statistical fault‑injection technique for FP32 single‑bit faults in deep neural network activations and weights. It partitions each layer’s value distribution into intervals with similar expected bit‑flip behavior using regression trees, then allocates injections across these intervals based on their relevance for failure‑rate estimation. This stratified approach preserves target confidence and error margins while dramatically reducing the required injection budget—up to 72.1× for activations and 11.2× for weights compared to existing baselines.
By Noam Bires, Marcello Traiola, Angeliki Kritikakou, Elisa Fromont
arXiv:2607. 12868v1 Announce Type: cross Abstract: Deep learning systems often fail due to subtle implementation faults that alter training behavior.
By Sigma Jahan
arXiv:2607. 11193v1 Announce Type: cross Abstract: To ensure the overall quality of AI-enabled software, not only traditional software components but also AI components need to be tested and repaired.
By Yuta Ishimoto, Paolo Arcaini, Fuyuki Ishikawa, Masanari Kondo, Naoyasu Ubayashi, Yasutaka Kamei
arXiv:2608. 04173v1 Announce Type: new Abstract: Deep neural networks (DNNs) deployed on resource-constrained neuromorphic hardware face three concurrent challenges: the need for model compression through pruning, vulnerability to adversarial input perturbations, and susceptibility to hardware-induced weight faults such as stuck-at-zero errors.
By Manali Dangarikar, Cory Merkel
arXiv:2603. 22770v2 Announce Type: replace-cross Abstract: The deployment of deep neural networks (DNNs) in safety-critical edge environments necessitates robustness against hardware-induced bit-flip errors.
By Alan T. L. Bacellar, Sathvik Chemudupati, Shashank Nag, Allison Seigler, Priscila M. V. Lima, Felipe M. G. Fran\c{c}a, Lizy K. John
arXiv:2606. 06536v1 Announce Type: cross Abstract: Automated defect detection in high-voltage transmission-line insulators remains challenging due to severe class imbalance, large scale variation, and the small spatial extent of defect instances in Unmanned Aerial Vehicle (UAV) imagery.
By Malak Allam, Khaled Shaban, Ali Hamdi
The paper introduces REQAP, a reliability‑aware quantized weight packing technique for systolic‑array DNN accelerators. It uses a sensitivity‑driven mixed‑precision quantization to assign layer‑wise bit‑widths, a deterministic register‑level packing strategy for SIMD‑within‑a‑register execution, and selective bit‑level protection that replicates critical MSBs into unused register space. Experiments on AlexNet, VGG‑11, and ResNet‑18 show up to 62% memory reduction, 56% fewer MAC operations, and improved accuracy resilience under fault injection compared to baseline and fully protected models.
By Mahdi Taheri, Samira Nazari, Mubassher Ansari, Ali Azarpeyvand, Mohsen Afsharchi, Maksim Jenihhin, Christian Herglotz
arXiv:2609.16742v1 Announce Type: cross
Abstract: Convolutional Neural Networks (CNNs) are increasingly deployed in safety-critical edge applications, where soft errors can silently corrupt inference...
By Kyrylo Nazarevych, Mohammad Hasan Ahmadilivani, Krister Kaldre, Davide Bertozzi, Jaan Raik
arXiv:2608. 04035v1 Announce Type: cross Abstract: The wide adoption of Vision Transformers (ViTs) in safety-critical applications raises reliability concerns related to hardware faults.
By Mohammad Hasan Ahmadilivani, Sven-Markus Loorits, Jaan Raik
arXiv:2606. 17471v1 Announce Type: new Abstract: Traditional CPU, GPU, and NPU architectures are increasingly limited by the von Neumann bottleneck.
By Ching-Yi Lin, Shamik Kundu, Arnab Raha, Sahil Shah
arXiv:2609.22631v1 Announce Type: new
Abstract: Accurate automated interpretation of electrocardio- grams (ECGs) is essential for early detection of cardiac condi- tions such as myocardial infarction...
By Mohammad Sadman Tahsin, Haitham Y. Adarbah, Afzel Noore
HiFiNet is a hierarchical fault identification framework for Wireless Sensor Networks that uses edge-based LSTM stacked autoencoders for initial temporal feature extraction and a Graph Attention Network to aggregate neighboring node information for refined classification. The approach captures both local temporal patterns and network-wide spatial dependencies, leading to higher accuracy, F1-score, and precision compared to existing methods. Experiments on synthetic datasets derived from the Intel Lab Dataset and NASA's MERRA-2 reanalysis data demonstrate HiFiNet’s robustness and its ability to balance diagnostic performance with energy efficiency.
By Nguyen Tri Nghia, Nguyen Van Son, Nguyen Thi Hanh