生成式 AI
AI 图像生成、视频生成、音乐创作等 AIGC 领域最新动态。
IntroductionDMD-augmented Unpaired Neural Schr\"odinger Bridge for Ultra-Low Field MRI Enhancement
Researchers developed an AI framework that translates ultra-low-field (64 mT) brain MRI scans to resemble high-fidelity ...
IntroductionDMD-augmented Unpaired Neural Schr\"odinger Bridge for Ultra-Low Field MRI Enhancement
Researchers developed a novel AI framework that enhances ultra-low-field (64 mT) brain MRI scans to resemble high-qualit...
IntroductionDMD-augmented Unpaired Neural Schr\"odinger Bridge for Ultra-Low Field MRI Enhancement
Researchers developed an AI framework using a Neural Schrödinger Bridge and frozen 3T diffusion model to enhance ultra-l...
JANUS: Structured Bidirectional Generation for Guaranteed Constraints and Analytical Uncertainty
JANUS (Joint Ancestral Network for Uncertainty and Synthesis) is a novel synthetic data generation framework that solves...
JANUS: Structured Bidirectional Generation for Guaranteed Constraints and Analytical Uncertainty
JANUS (Joint Ancestral Network for Uncertainty and Synthesis) is a novel synthetic data generation framework that solves...
JANUS: Structured Bidirectional Generation for Guaranteed Constraints and Analytical Uncertainty
JANUS (Joint Ancestral Network for Uncertainty and Synthesis) is a novel synthetic data generation framework that solves...
JANUS: Structured Bidirectional Generation for Guaranteed Constraints and Analytical Uncertainty
JANUS (Joint Ancestral Network for Uncertainty and Synthesis) is a novel framework that solves the synthetic data Quadri...
Order Is Not Layout: Order-to-Space Bias in Image Generation
Researchers have identified a systematic Order-to-Space Bias (OTS) in AI image generators like Stable Diffusion and DALL...
Order Is Not Layout: Order-to-Space Bias in Image Generation
Researchers have identified Order-to-Space Bias (OTS), a systematic flaw where AI image generators like Stable Diffusion...
Order Is Not Layout: Order-to-Space Bias in Image Generation
Researchers have identified a systematic Order-to-Space Bias (OTS) in modern AI image generators like Stable Diffusion, ...
Order Is Not Layout: Order-to-Space Bias in Image Generation
Researchers have identified a systematic Order-to-Space Bias (OTS) in AI image generation models like Stable Diffusion a...
Generalization Properties of Score-matching Diffusion Models for Intrinsically Low-dimensional Data
A new theoretical analysis demonstrates that score-based diffusion models can learn data distributions with convergence ...
Generalization Properties of Score-matching Diffusion Models for Intrinsically Low-dimensional Data
New theoretical research provides the first rigorous statistical guarantees for score-based diffusion models, demonstrat...
Generalization Properties of Score-matching Diffusion Models for Intrinsically Low-dimensional Data
Theoretical research establishes rigorous statistical convergence guarantees for score-based diffusion models, demonstra...
Generalization Properties of Score-matching Diffusion Models for Intrinsically Low-dimensional Data
Recent theoretical work establishes that score-based diffusion models achieve statistical convergence rates that scale w...
Generalization Properties of Score-matching Diffusion Models for Intrinsically Low-dimensional Data
A new theoretical analysis establishes that score-based diffusion models achieve convergence rates that depend on the in...
Error as Signal: Stiffness-Aware Diffusion Sampling via Embedded Runge-Kutta Guidance
Embedded Runge-Kutta Guidance (ERK-Guid) is a novel diffusion model sampling method that uses solver-induced local trunc...
Error as Signal: Stiffness-Aware Diffusion Sampling via Embedded Runge-Kutta Guidance
Researchers introduced Embedded Runge-Kutta Guidance (ERK-Guid), a novel method that addresses solver-induced local trun...
Error as Signal: Stiffness-Aware Diffusion Sampling via Embedded Runge-Kutta Guidance
Embedded Runge-Kutta Guidance (ERK-Guid) is a novel diffusion model sampling method that uses the local truncation error...
Error as Signal: Stiffness-Aware Diffusion Sampling via Embedded Runge-Kutta Guidance
Embedded Runge-Kutta Guidance (ERK-Guid) is a novel diffusion model sampling method that repurposes numerical solver err...
Error as Signal: Stiffness-Aware Diffusion Sampling via Embedded Runge-Kutta Guidance
Embedded Runge-Kutta Guidance (ERK-Guid) is a novel guidance method for diffusion models that uses solver-induced error ...
PhyPrompt: RL-based Prompt Refinement for Physically Plausible Text-to-Video Generation
PhyPrompt is a reinforcement learning framework that automatically refines text prompts to generate physically plausible...
PhyPrompt: RL-based Prompt Refinement for Physically Plausible Text-to-Video Generation
PhyPrompt is a novel AI framework that uses reinforcement learning to automatically refine text prompts for generating p...
PhyPrompt: RL-based Prompt Refinement for Physically Plausible Text-to-Video Generation
PhyPrompt is a reinforcement learning framework that automatically refines text prompts to generate physically plausible...
PhyPrompt: RL-based Prompt Refinement for Physically Plausible Text-to-Video Generation
PhyPrompt is a novel reinforcement learning framework that automatically refines text prompts to generate physically pla...
Phys4D: Fine-Grained Physics-Consistent 4D Modeling from Video Diffusion
Google DeepMind's Phys4D is a novel AI pipeline that systematically injects physical consistency into video diffusion mo...
Phys4D: Fine-Grained Physics-Consistent 4D Modeling from Video Diffusion
Phys4D is a novel pipeline that addresses the physical inconsistency of video diffusion models by training them to under...
Phys4D: Fine-Grained Physics-Consistent 4D Modeling from Video Diffusion
Google DeepMind's Phys4D is a novel three-stage training pipeline that transforms appearance-driven video diffusion mode...
Phys4D: Fine-Grained Physics-Consistent 4D Modeling from Video Diffusion
Phys4D is a novel pipeline that addresses physical inconsistencies in video diffusion models through a three-stage train...
Phys4D: Fine-Grained Physics-Consistent 4D Modeling from Video Diffusion
Phys4D is a novel pipeline developed by UC Berkeley researchers that transforms appearance-driven video diffusion models...
Beyond Pixel Histories: World Models with Persistent 3D State
PERSIST is a groundbreaking world model architecture that shifts interactive video generation from learning 2D patterns ...
PRIVATEEDIT: A Privacy-Preserving Pipeline for Face-Centric Generative Image Editing
PRIVATEEDIT is a privacy-preserving pipeline for face-centric generative AI editing that prevents biometric data from be...
ByteDance’s AI Ambitions Are Being Hampered by Compute Restraints and Copyright Concerns
ByteDance's Seedance 2.0 AI video generator, positioned as a competitor to OpenAI's Sora, has encountered significant op...
ByteDance’s AI Ambitions Are Being Hampered by Compute Restraints and Copyright Concerns
ByteDance's Seedance 2.0 AI video generation model is experiencing significant operational strain due to overwhelming us...
Cryo-SWAN: the Multi-Scale Wavelet-decomposition-inspired Autoencoder Network for molecular density representation of molecular volumes
Cryo-SWAN is a voxel-based variational autoencoder designed for 3D molecular density data, such as cryo-electron microsc...
Cryo-SWAN: the Multi-Scale Wavelet-decomposition-inspired Autoencoder Network for molecular density representation of molecular volumes
Cryo-SWAN is a novel voxel-based variational autoencoder designed specifically for 3D molecular density volumes from cry...
Cryo-SWAN: the Multi-Scale Wavelet-decomposition-inspired Autoencoder Network for molecular density representation of molecular volumes
Cryo-SWAN is a novel voxel-based variational autoencoder specifically designed for 3D molecular density volumes from cry...
Generative AI in Managerial Decision-Making: Redefining Boundaries through Ambiguity Resolution and Sycophancy Analysis
A new study reveals that Generative AI serves as a cognitive scaffold for identifying ambiguities in business decision-m...
Generative AI in Managerial Decision-Making: Redefining Boundaries through Ambiguity Resolution and Sycophancy Analysis
New research establishes a framework for evaluating generative AI in managerial decision-making, revealing its strengths...
Grammarly Is Offering ‘Expert’ AI Reviews From Your Favorite Authors—Dead or Alive
Superhuman (formerly OthersideAI) has launched an AI writing tool that critiques user text by emulating the styles of re...
Apple Music adds optional labels for AI songs and visuals
Apple Music is launching a voluntary 'Transparency Tags' system for AI-generated content, allowing rights holders to sel...
How 1,000+ customer calls shaped a breakout enterprise AI startup
Narada, an AI platform for music and sound generation, developed its product strategy through 1,000+ customer calls, pra...
Netflix buys Ben Affleck’s AI filmmaking company InterPositive
Netflix has acquired InterPositive, Ben Affleck's AI filmmaking company that developed a specialized AI model for post-p...
Netflix buys Ben Affleck’s AI filmmaking company InterPositive
Netflix has acquired InterPositive, Ben Affleck's AI filmmaking company that developed a novel post-production tool. Unl...
Introducing Modular Diffusers - Composable Building Blocks for Diffusion Pipelines
Modular diffusers are composable building blocks designed for diffusion pipelines, which are frameworks for AI image gen...
Seedance2.0生成视频价格公布，生成视频一秒1块钱
字节跳动旗下火山引擎正式公布了其视频生成模型Seedance2.0的商用定价。服务分为含视频输入的编辑模式（28元/百万tokens）和不含视频输入的纯生成模式（46元/百万tokens）。根据官方数据，生成一段15秒的标准视频约消耗30....
Physically Valid Biomolecular Interaction Modeling with Gauss-Seidel Projection
Researchers have developed a differentiable AI module that enforces strict steric feasibility in biomolecular interactio...
Physically Valid Biomolecular Interaction Modeling with Gauss-Seidel Projection
A novel differentiable Gauss-Seidel projection module enforces physical constraints in AI-generated biomolecular structu...
Physically Valid Biomolecular Interaction Modeling with Gauss-Seidel Projection
A novel AI module integrates a differentiable Gauss-Seidel projection to enforce physical steric constraints in biomolec...
Higher Gauge Flow Models
Higher Gauge Flow Models are a novel class of generative AI models that extend ordinary Gauge Flow Models by incorporati...
Higher Gauge Flow Models
Higher Gauge Flow Models represent a novel class of generative artificial intelligence that incorporates L∞-algebra stru...
Higher Gauge Flow Models
Higher Gauge Flow Models represent a novel generative AI architecture that extends traditional Gauge Flow Models by inco...
Higher Gauge Flow Models
Higher Gauge Flow Models represent a groundbreaking class of generative AI that extends ordinary Gauge Flow Models by in...
Higher Gauge Flow Models
Higher Gauge Flow Models represent a novel class of generative AI that incorporates advanced geometric structures like L...
Function-Space Decoupled Diffusion for Forward and Inverse Modeling in Carbon Capture and Storage
The Function-space Decoupled Diffusion Posterior Sampling (Fun-DDPS) framework combines generative diffusion models with...
Function-Space Decoupled Diffusion for Forward and Inverse Modeling in Carbon Capture and Storage
The Function-space Decoupled Diffusion Posterior Sampling (Fun-DDPS) framework is a novel generative AI method that comb...
Function-Space Decoupled Diffusion for Forward and Inverse Modeling in Carbon Capture and Storage
The Function-space Decoupled Diffusion Posterior Sampling (Fun-DDPS) framework is a novel AI method that combines genera...
Continual Unlearning for Text-to-Image Diffusion Models: A Regularization Perspective
A new study reveals that text-to-image diffusion models experience 'utility collapse' during continual unlearning, where...
Continual Unlearning for Text-to-Image Diffusion Models: A Regularization Perspective
New research reveals that text-to-image diffusion models suffer from rapid utility collapse when processing sequential u...
Fine-Tuning Diffusion Models via Intermediate Distribution Shaping
Researchers introduced P-GRAFT, a novel fine-tuning framework for diffusion models that shapes intermediate probability ...