ResNet-34
PulseAugur coverage of ResNet-34 — every cluster mentioning ResNet-34 across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
AI models often sound confident even when wrong, a fluency vs. accuracy issue
A common failure mode in AI models is overconfidence, where they sound certain even when incorrect. This phenomenon, observed in models like ResNet-34 and across various architectures as noted by Guo et al. (2017), stem…
-
New research explores functional modularity in neural networks and efficient SNN quantization
Two new research papers explore advancements in neural network architectures and efficiency. The first paper, "Sparsity-driven emergence of functional modularity in multitask neural networks," demonstrates how sparsity-…
-
New FETALFusion model advances fetal head segmentation in ultrasound
Researchers have developed FETALFusion, a novel dual-path CNN-Mamba model designed for accurate fetal head segmentation in ultrasound images. This model integrates state-space modeling at high-resolution stages and util…
-
New integer-only training method slashes neural network memory use
Researchers have developed a novel training algorithm for deep neural networks that exclusively uses integer arithmetic, eliminating the need for floating-point operations in any stage of the training process. This meth…
-
Contrast enhancement boosts cervical cancer classification accuracy in CNNs
A new research paper published on arXiv explores the effectiveness of image preprocessing techniques for improving cervical cancer classification using deep learning models. The study evaluated two methods, a PMD filter…
-
Compact robot policy achieves high performance with fine-grained visual representations
Researchers have developed CoRP, a compact robot policy that achieves high performance with significantly fewer parameters than existing systems. The study emphasizes that the effectiveness of multi-task manipulation po…
-
Compact robot policy CoRP achieves high performance with efficient visual representation
Researchers have developed CoRP (Compressed Representation Policy), a compact robot manipulation policy that achieves high performance with significantly fewer parameters and no large vision-language models or video-gen…
-
GeoAI framework maps invasive trees threatening Hawaii infrastructure
A new study introduces a GeoAI framework to map the exposure of infrastructure to invasive Albizia trees in Hawaii. The framework utilizes a U-Net convolutional neural network with a ResNet-34 backbone to identify and m…
-
ResNets implicitly route by varying residual interaction experts
A new research paper explores the internal workings of Residual Networks (ResNets), suggesting they implicitly perform a form of soft routing. By analyzing ResNet-18 and ResNet-34 models trained on ImageNet, the study f…
-
Research paper links CNN topology to trainability, not just layer count
A new research paper explores the relationship between the nominal depth of convolutional neural networks (CNNs) and their trainability, introducing the concept of "effective depth." The study found that while nominal d…
-
New framework enhances needle-tip localization in ultrasound videos
Researchers have developed STUNet-Fusion, a novel spatiotemporal framework designed to improve needle-tip localization in ultrasound videos. This method addresses challenges such as weak or discontinuous needle visibili…
-
New dataset enhances trajectory scoring for autonomous driving
Researchers have developed a new training dataset designed to improve the performance of learned trajectory scoring models in autonomous driving systems. This dataset focuses on providing more informative supervision by…
-
Federated learning framework improves bladder cancer segmentation and classification
Researchers have developed a federated multi-task learning framework to improve bladder tumor segmentation and muscle-invasive bladder cancer (MIBC) classification using T2-weighted MRI. The proposed Swin Hybrid model, …
-
Deep learning model accurately detects Christmas tree plantations in aerial imagery
Researchers have developed a deep learning framework to accurately identify Christmas tree plantations using high-resolution aerial imagery. The study, focusing on the French Morvan region, addresses the unique challeng…
-
New SHANG++ method enhances AI model training robustness against noise
Researchers have developed two new accelerated stochastic gradient descent methods, SHANG and SHANG++, designed to improve stability and convergence under multiplicative noise. SHANG is a semi-implicit discretization th…
-
GeoAI framework automates building footprint validation for GIS databases
Researchers have developed a GeoAI framework to automatically validate and purify building footprint data extracted from high-resolution imagery. This framework uses spatial feature engineering and machine learning clas…
-
Perforated AI enhances ResNet-18 to match ResNet-34 performance
Perforated AI has developed a technique to enhance the performance of the ResNet-18 model, bringing its accuracy on par with the larger ResNet-34. This method, inspired by biological processes, achieves improved results…
-
New research explores advanced federated learning techniques · 10 sources tracked
Multiple research papers published on arXiv in August 2026 introduce novel approaches to enhance federated learning (FL) and decentralized FL. These methods address challenges such as modality missingness in multimodal …
-
GeoAI tutorial details building footprint extraction using U-Net, DINO, SAM, and Mask R-CNN
This tutorial details a GeoAI workflow for extracting building footprints from aerial imagery using a combination of deep learning models. It covers setting up the geospatial environment, training a U-Net model with a R…
-
New StoMPP method improves binary neural network training
Researchers have introduced StoMPP (Stochastic Masked Partial Progressive Binarization), a novel training method for binary neural networks (BNNs) that avoids the accuracy degradation typically seen with deeper networks…