ResNet50
PulseAugur coverage of ResNet50 — every cluster mentioning ResNet50 across labs, papers, and developer communities, ranked by signal.
- 2026-06-17 research_milestone A new two-stage fine-tuning method for ResNet50 was published on arXiv for improved melanoma detection. source
6 day(s) with sentiment data
-
New RLGPA method uses reinforcement learning for targeted adversarial attacks
Researchers have developed RLGPA, a novel method that uses reinforcement learning to guide the process of creating sparse adversarial examples for AI models. This approach dynamically schedules and adjusts the granulari…
-
Deep learning models benchmarked for breast cancer diagnosis
Researchers have evaluated ten deep learning models, including CNNs like DenseNet121 and transformer models such as ViT and Swin Transformer, for their effectiveness in diagnosing breast cancer from medical images. Dens…
-
Deep Regression Method Extracts Medical Data from UK Biobank MRIs
Researchers have developed a novel methodology using targeted deep regression to extract numerical properties from whole-body water-fat MRI scans. This approach employs a ResNet50-style network on collages representing …
-
ResNet50 outperforms Vision Transformer in efficiency for GI disease classification
A new benchmark study evaluated ResNet50 and Vision Transformer (ViT-B/16) for classifying gastrointestinal diseases from endoscopic images. While both models achieved high accuracy (around 92%) in cross-validation, Res…
-
New UTA method optimizes learning rate decay for AI models
Researchers from the University of Texas at Austin have developed a new learning rate decay method called Universal Training Adaptive (UTA) LRD. This method aims to improve model training by being adaptive to the traini…
-
New TWA Method Enhances Out-of-Distribution Detection in AI Models
Researchers have developed a new method called Rank-1 Weight-Corrected Trainable Weight Averaging (TWA) to improve out-of-distribution (OOD) detection in AI models. This technique builds upon existing TWA methods by inc…
-
New FSVL framework enhances fish identification with dual-path AI
Researchers have developed FSVL, a novel framework that combines fast and slow pathways for identifying fish species in the Middle Yangtze River Basin. The fast pathway, using ResNet50, achieves high preliminary accurac…
-
AI models' concept accuracy audited, revealing potential benchmark flaws
A new research paper published on arXiv investigates the accuracy of Concept Bottleneck Models (CBMs) and Concept Embedding Models (CEMs) on the CUB-200-2011 dataset. The study reveals that while CBMs promise human-insp…
-
AI model accurately classifies retinal development across OCT devices
Researchers have developed and validated an unsupervised domain adaptation model to accurately classify normal versus abnormal retinal development from optical coherence tomography (OCT) scans. This model, based on a Re…
-
New research explores Corrective Power in pairwise AUC optimization
A new research paper explores the concept of "Corrective Power" in optimizing pairwise AUC (Area Under the Curve) for machine learning models, particularly in scenarios with imbalanced classes. The study identifies that…
-
New deep learning patching method balances performance and training time
A new paper introduces Stride Independent Patching (SSP), a semi-automatic method for positioning image patches in deep learning. Unlike automatic stride-dependent methods, SSP relies on user input to define patch locat…
-
New research highlights "representation risk" in image encoders
A new research paper introduces the concept of "representation risk" in pretrained image encoders, demonstrating that different encoders can lead to significantly different outcomes in downstream prediction tasks. The s…
-
New AI framework CirrGuide improves liver cirrhosis segmentation and classification
Researchers have developed CirrGuide, a novel deep learning framework designed to segment liver cirrhosis from T2-weighted MRI scans and classify its severity. The framework employs a cascaded approach, first predicting…
-
Bird ID models: Resolution vs. Architecture trade-offs on edge devices
A new study investigates the optimal input resolution for bird species identification models, particularly for edge devices like the NVIDIA Jetson Orin Nano. Researchers found that model architecture significantly impac…
-
Land Art Serves as AI-Powered Climate Indicator
Researchers have developed a novel method to use Robert Smithson's 1970 land artwork, Spiral Jetty, as a climate sensor by analyzing satellite imagery. By examining 1,744 image chips from 1984 to 2025, they extracted co…
-
CNNs Compared for Melanoma Detection Across Image Types
A new research paper evaluates the effectiveness of several pre-trained convolutional neural networks (CNNs) for melanoma detection using both dermatoscopic and histopathological images. The study utilized datasets such…
-
New method uses defect masks for spatial supervision in AI inspection
Researchers have developed a novel method for defect localization in industrial inspection by repurposing ground-truth defect masks as spatial supervision signals during model training. This approach enhances the abilit…
-
New framework enhances multimodal emotion recognition using attention-based fusion
Researchers have developed a new framework for multimodal emotion recognition, integrating audio and visual data. The audio component uses Wav2Vec2, MFCCs, and acoustic descriptors processed by a BiLSTM, while the video…
-
Medical foundation models enhance brain MRI contrast dose simulation
Researchers have developed a new method for simulating brain MRI contrast doses by utilizing features from medical foundation models as a perceptual loss. This approach aims to improve the accuracy of image synthesis co…
-
Deep learning system accurately identifies Bangladeshi mango varieties
Researchers have developed a deep learning-powered web system to identify Bangladeshi mango varieties, addressing the challenge of distinguishing similar cultivars. The system utilizes three fine-tuned CNN architectures…