ImageNet
PulseAugur coverage of ImageNet — every cluster mentioning ImageNet across labs, papers, and developer communities, ranked by signal.
- used by Deep Neural Networks 90%
- used by arXiv 70%
- used by CIFAR-10 70%
- used by Diffusion Models 70%
- used by residual neural network 70%
- instance of Diffusion Models 70%
- used by CIFAR-100 70%
- used by vision transformer 70%
- used by COCO 70%
- used by ConvNeXt 70%
- instance of magazine 70%
- used by Vít 70%
19 day(s) with sentiment data
-
New FMGP method enhances deep learning uncertainty estimation
Researchers have developed a new method called fixed-mean Gaussian Processes (FMGP) for estimating uncertainty in pre-trained deep neural networks. This approach fixes the Gaussian Process posterior mean to the DNN's ou…
-
SEAOTTER framework boosts robotics compression with JPEG compatibility
Researchers have developed SEAOTTER, a novel compression framework designed for cloud robotics that addresses bandwidth and compute limitations. This system combines a learned latent representation with the widely compa…
-
Arctic remote sensing model achieves high-resolution analysis
Researchers have developed a new foundation model specifically for analyzing very high-resolution satellite imagery of the Arctic. This model, trained using a masked autoencoder approach on a curated dataset of approxim…
-
New XAI methods detect absent concepts in neural networks
Researchers have introduced new methods to improve explainable AI (XAI) by identifying when a neuron's activation signifies the absence of a concept, rather than its presence. Current XAI techniques often struggle to de…
-
Generative Drifting identified as Score Matching in new research
A new paper proposes that Generative Drifting, a method for one-step image generation, is fundamentally a form of score matching. The research reveals that under specific conditions, the drift operator in this technique…
-
New framework offers stable, provable guarantees for AI circuit discovery
Researchers have developed a new framework called Certified Circuits to improve the reliability of identifying mechanistic circuits within neural networks. This method provides provable stability guarantees, ensuring th…
-
SRC-Flow uses compact representations for image generation
Researchers have developed SRC-Flow, a novel method for image generation using normalizing flows. This approach addresses the challenge of high-dimensional representations in visual data by first compressing features in…
-
DIVER framework enhances dataset distillation with diffusion models
Researchers have introduced DIVER, a novel dual-stage framework for dataset distillation that aims to improve privacy and learning efficiency. Unlike previous single-stage methods that overfit to specific architectures,…
-
DASH framework generates stealthy adversarial AI examples
Researchers have developed DASH, a meta-attack framework designed to create adversarial examples for AI models that are both effective at causing misclassification and visually imperceptible. This framework strategicall…
-
Network pruning impacts GoogLeNet performance and interpretability
Researchers investigated how network pruning affects the performance and interpretability of GoogLeNet on ImageNet. They applied various pruning techniques and retraining strategies, finding that performance could be ma…
-
New CAFD method uses VLMs for efficient DNN fault detection
Researchers have developed a new method called Concept-Aware Fault Detection (CAFD) to identify errors in Deep Neural Networks (DNNs). CAFD integrates various data sources, including a novel "Concept Failure Ratio" deri…
-
New framework uses approximate latent structure for certifiable classifier robustness
Researchers have developed a new framework to create certifiably robust deep learning classifiers by leveraging the latent structure within data representations. Their method proves that even approximate Gaussian mixtur…
-
New attack framework targets AI models with theoretical guarantees
Researchers have developed a new framework for adversarial attacks on AI models, focusing on hard-label black-box scenarios where only the top prediction is accessible. Their approach introduces a novel zero-query initi…
-
Vision Transformers improved with selective token interaction
Researchers have identified a phenomenon called "semantic diffusion" that degrades the performance of Vision Transformers (ViTs) in dense prediction tasks over time. This occurs when global semantic information spreads …
-
New RBDC protocol slashes vision model training costs by 30%
Researchers have developed a new training protocol called RBDC to make training large vision models more resource-efficient. This method involves recursively coupling independently trained, narrower models in a paramete…
-
DINOv3 vs ImageNet: Transfer learning for industrial vision tasks
A new research paper explores the effectiveness of transfer learning for industrial visual inspection tasks. The study compares DINOv3, a self-supervised model, against traditional ImageNet pretraining for RGB and X-ray…
-
ConvNeXt-FD model enhances biomedical image segmentation
Researchers have developed ConvNeXt-FD, a new deep learning model for segmenting biomedical images. This model utilizes a U-Net-like structure with a ConvNeXt backbone and incorporates a novel loss function that include…
-
New Gaussian Mixture Model improves DDIM sampling quality
Researchers have developed a new method to improve the sampling process in Denoising Diffusion Implicit Models (DDIM). Their approach utilizes a Gaussian Mixture Model (GMM) as the reverse transition operator, which mat…
-
New MDSE attack fools Spiking Neural Networks and traditional models
Researchers have developed a new adversarial attack method called Mixed Dynamic Spiking Estimation (MDSE) specifically for Spiking Neural Networks (SNNs). This attack demonstrates that the effectiveness of white-box adv…
-
TextTeacher uses language embeddings to boost vision model accuracy
Researchers have developed TextTeacher, a novel method to enhance vision model performance by leveraging language embeddings. This technique injects text information from image captions into the training process of visi…