Denoising Diffusion Probabilistic Models
PulseAugur coverage of Denoising Diffusion Probabilistic Models — every cluster mentioning Denoising Diffusion Probabilistic Models across labs, papers, and developer communities, ranked by signal.
7 day(s) with sentiment data
-
Diffusion inpainting generates healthy mammograms to boost cancer detection accuracy
Researchers have developed a novel counterfactual data augmentation technique using diffusion inpainting to improve mammography classification for computer-aided diagnosis (CAD) systems. This method generates realistic …
-
New hybrid model fuses patch detection and diffusion for semiconductor anomaly detection
Researchers have developed a novel hybrid anomaly detection framework for semiconductor wafer bin maps, combining a patch-based detector (EfficientAD) with a diffusion-driven generative model (DDPM). This fused approach…
-
New spectral diffusion strategy improves image generation quality
Researchers have developed a novel energy-adaptive noise scheduling and whitening strategy for spectral diffusion models used in image generation. This new approach, detailed in a paper submitted to arXiv, aims to impro…
-
New hybrid AI framework enhances pharmaceutical identification via Raman spectroscopy
Researchers have developed HyMLRaman, a novel framework for identifying pharmaceutical compounds using Raman spectroscopy. This hybrid approach integrates deep learning for feature extraction, generative models for data…
-
New method efficiently attributes training data for diffusion models
Researchers have developed a new method called TID (Training-data Influence via score Discrepancy) to efficiently attribute training data for diffusion models. This approach estimates influence without costly per-sample…
-
New SANI framework enhances diffusion models with pixel-level noise injection
Researchers have introduced Spatially Adaptive Noise Injection (SANI), a new framework for diffusion models that dynamically adjusts noise application on a per-pixel basis. Unlike traditional methods that apply noise un…
-
Flow Matching Model Predicts Aircraft Trajectories with High Accuracy
Researchers have developed FlowATC, a novel architecture for predicting aircraft trajectories using flow matching techniques. Trained on over a million Automatic Dependent Surveillance-Broadcast (ADS-B) trajectory windo…
-
New framework links DDPM score matching to parameter and density estimation
A new research paper introduces a framework that connects score estimation in denoising diffusion probabilistic models (DDPMs) to parameter and density estimation tasks. This framework demonstrates that DDPM score match…
-
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 research optimizes LoRA fine-tuning for AI models · 2 papers
Two new research papers explore optimizations for Low-Rank Adaptation (LoRA) fine-tuning. The first paper investigates the trade-offs between LoRA rank, model quality, and computational cost, finding that moderate ranks…
-
Diffusion models adapted for discrete tasks and maximum entropy generation
Researchers are exploring novel diffusion model techniques to improve performance on discrete tasks. One approach involves modifying the sampling process to prevent early errors from persisting, significantly boosting a…
-
Diffusion models generate realistic eye-gaze trajectories for research and biometrics
Researchers have developed new diffusion models for generating realistic eye-gaze trajectories, addressing the challenge of expensive and privacy-constrained data collection. One model focuses on generating raw 2D gaze …
-
Mode connectivity demonstrated in generative and contrastive AI models
Researchers have demonstrated mode connectivity in generative and contrastive models, extending previous findings that were limited to classifiers. By developing an architecture-aware connection algorithm tailored for D…
-
Diffusion models enhance medical image segmentation and quality assurance · 2 sources tracked
Researchers are exploring the use of diffusion models for medical image segmentation, a process crucial for radiotherapy planning. One study proposes using unsupervised diffusion models to pretrain encoders for segmenta…
-
Diffusion models research tackles outliers, efficiency, and theory · 10 sources tracked
Recent research explores advancements in diffusion models, focusing on improving their robustness, efficiency, and theoretical understanding. Papers address challenges like outlier data in inverse problems, scaling rein…
-
Synthetic data boosts AI analysis of Ukraine war-damaged fields
Researchers have developed a method using synthetic data augmentation to improve the analysis of battle-damaged agricultural fields in Ukraine. By training generative models like Generative Adversarial Networks (GANs) a…
-
RadioVIL framework enhances 6G radio maps with anomaly detection for vehicle localization
Researchers have developed RadioVIL, a novel two-stage framework for high-precision radio map construction essential for 6G Integrated Sensing and Communication (ISAC) applications. Unlike previous methods that smooth o…
-
AI agents automate PET image denoising using VLM and LLM
Researchers have developed a novel multi-agent system that leverages vision-language models (VLMs) and large language models (LLMs) to automate and enhance the denoising of Positron Emission Tomography (PET) images. Thi…
-
New FiRe framework speeds up AI visual counterfactual explanations
Researchers have developed FiRe, a novel framework for generating visual counterfactual explanations in AI models. This method refines images at a fixed noise level, unlike previous approaches that followed a longer, va…
-
AI system PHOENIX enables autonomous satellite self-healing
A new research paper introduces PHOENIX, a system designed to extend the operational lifespan of CubeSats by enabling them to perform predictive self-healing and autonomous recovery. The system utilizes a fine-tuned sma…