行业应用
AI 在医疗、金融、教育、制造等各行业的落地应用与案例分析。
CAM-LDS: Cyber Attack Manifestations for Automatic Interpretation of System Logs and Security Alerts
The Cyber Attack Manifestation Log Data Set (CAM-LDS) is a publicly available benchmark dataset designed to advance AI-d...
Understanding Sources of Demographic Predictability in Brain MRI via Disentangling Anatomy and Contrast
A novel disentangled representation learning framework separates anatomical information from acquisition-dependent contr...
Understanding Sources of Demographic Predictability in Brain MRI via Disentangling Anatomy and Contrast
A novel disentangled representation learning framework reveals that demographic predictability in brain MRI scans origin...
Understanding Sources of Demographic Predictability in Brain MRI via Disentangling Anatomy and Contrast
A novel disentangled representation learning framework reveals that demographic predictability in brain MRI scans—includ...
Understanding Sources of Demographic Predictability in Brain MRI via Disentangling Anatomy and Contrast
A novel disentangled representation learning framework reveals that demographic predictability in brain MRI scans is pri...
End-to-end event reconstruction for precision physics at future colliders
A new end-to-end machine learning framework for global event reconstruction at future particle colliders maps raw detect...
End-to-end event reconstruction for precision physics at future colliders
A novel end-to-end machine learning framework for global event reconstruction at future particle colliders achieves 10-2...
End-to-end event reconstruction for precision physics at future colliders
A novel AI-based end-to-end event reconstruction method using geometric algebra transformer networks and object condensa...
End-to-end event reconstruction for precision physics at future colliders
A novel end-to-end AI model for particle physics event reconstruction uses geometric algebra transformer networks and ob...
End-to-end event reconstruction for precision physics at future colliders
A new end-to-end machine learning framework for global event reconstruction at particle colliders directly maps raw dete...
SaFeR: Safety-Critical Scenario Generation for Autonomous Driving Test via Feasibility-Constrained Token Resampling
SaFeR is a novel AI framework for generating safety-critical autonomous driving test scenarios that balances adversarial...
The Empty Quadrant: AI Teammates for Embodied Field Learning
The academic paper 'Field Atlas: Reorienting AIED Toward Embodied, Dialogic Sensemaking in the Wild' proposes a radical ...
The Empty Quadrant: AI Teammates for Embodied Field Learning
The research paper 'Field Atlas: Reorienting AIED Toward Embodied, Dialogic Sensemaking' challenges the four-decade-old ...
The Empty Quadrant: AI Teammates for Embodied Field Learning
The Field Atlas framework represents a paradigm shift in artificial intelligence in education (AIED), moving from screen...
The Empty Quadrant: AI Teammates for Embodied Field Learning
Researchers from UC Berkeley and University of Washington propose Field Atlas, a novel AI framework that transforms arti...
The Empty Quadrant: AI Teammates for Embodied Field Learning
Researchers from UC Berkeley and University of Washington propose the Field Atlas framework, which repositions AI as an ...
Volumetric Directional Diffusion: Anchoring Uncertainty Quantification in Anatomical Consensus for Ambiguous Medical Image Segmentation
Volumetric Directional Diffusion (VDD) is a novel AI model for 3D medical lesion segmentation that quantifies uncertaint...
Volumetric Directional Diffusion: Anchoring Uncertainty Quantification in Anatomical Consensus for Ambiguous Medical Image Segmentation
Volumetric Directional Diffusion (VDD) is a novel AI model for 3D medical image segmentation that addresses ambiguous le...
STEM Faculty Perspectives on Generative AI in Higher Education
A study of 29 STEM faculty at a U.S. public university reveals varied engagement with generative AI tools like ChatGPT a...
STEM Faculty Perspectives on Generative AI in Higher Education
A focus group study of 29 STEM faculty at a large U.S. public university reveals that generative AI adoption is largely ...
STEM Faculty Perspectives on Generative AI in Higher Education
A study of 29 STEM faculty members reveals that generative AI adoption in higher education is primarily student-driven, ...
STEM Faculty Perspectives on Generative AI in Higher Education
A study of 29 STEM faculty at a U.S. public university reveals that generative AI adoption in higher education is primar...
STEM Faculty Perspectives on Generative AI in Higher Education
A study of 29 STEM faculty members reveals a spectrum of engagement with generative AI tools like ChatGPT and GitHub Cop...
TFWaveFormer: Temporal-Frequency Collaborative Multi-level Wavelet Transformer for Dynamic Link Prediction
TFWaveFormer is a novel Transformer architecture that integrates temporal-frequency analysis with multi-resolution wavel...
TFWaveFormer: Temporal-Frequency Collaborative Multi-level Wavelet Transformer for Dynamic Link Prediction
TFWaveFormer is a novel Transformer architecture that integrates temporal-frequency coordination with learnable multi-re...
Selecting Offline Reinforcement Learning Algorithms for Stochastic Network Control
A technical evaluation of offline reinforcement learning algorithms for stochastic wireless network control found Conser...
Selecting Offline Reinforcement Learning Algorithms for Stochastic Network Control
A comprehensive study evaluates offline reinforcement learning algorithms for stochastic wireless network control, findi...
Selecting Offline Reinforcement Learning Algorithms for Stochastic Network Control
A comprehensive benchmarking study reveals Conservative Q-Learning (CQL) as the most robust offline reinforcement learni...
Selecting Offline Reinforcement Learning Algorithms for Stochastic Network Control
A comprehensive benchmarking study reveals Conservative Q-Learning (CQL) as the most robust offline reinforcement learni...
Selecting Offline Reinforcement Learning Algorithms for Stochastic Network Control
A systematic evaluation of offline reinforcement learning algorithms in stochastic wireless environments reveals Conserv...
PatchDecomp: Interpretable Patch-Based Time Series Forecasting
PatchDecomp is a novel neural network-based time series forecasting model that balances predictive accuracy with interpr...
PatchDecomp: Interpretable Patch-Based Time Series Forecasting
PatchDecomp is a novel neural network method for time series forecasting that decomposes predictions into interpretable ...
PatchDecomp: Interpretable Patch-Based Time Series Forecasting
PatchDecomp is a novel neural network-based time series forecasting method that uniquely combines high predictive accura...
PatchDecomp: Interpretable Patch-Based Time Series Forecasting
PatchDecomp is a novel AI model for time series forecasting that bridges the accuracy-interpretability gap by decomposin...
PatchDecomp: Interpretable Patch-Based Time Series Forecasting
PatchDecomp is a novel neural network-based method for time series forecasting that uniquely combines high predictive ac...
A novel network for classification of cuneiform tablet metadata
Researchers have developed a novel convolution-inspired neural network specifically designed for classifying metadata fr...
A novel network for classification of cuneiform tablet metadata
Researchers have developed a novel deep learning architecture specifically designed to classify metadata from 3D scans o...
A novel network for classification of cuneiform tablet metadata
Researchers have developed a novel neural network architecture specifically designed to classify metadata from 3D point ...
A novel network for classification of cuneiform tablet metadata
Researchers have developed a novel convolution-inspired neural network architecture for classifying metadata from 3D sca...
A novel network for classification of cuneiform tablet metadata
Researchers have developed a novel convolution-inspired neural network architecture specifically designed to classify me...
Fairness Begins with State: Purifying Latent Preferences for Hierarchical Reinforcement Learning in Interactive Recommendation
Researchers from the University of Science and Technology of China have developed DSRM-HRL, a novel AI framework for int...
Fairness Begins with State: Purifying Latent Preferences for Hierarchical Reinforcement Learning in Interactive Recommendation
Researchers from the University of Science and Technology of China developed DSRM-HRL, a novel AI framework that address...
Fairness Begins with State: Purifying Latent Preferences for Hierarchical Reinforcement Learning in Interactive Recommendation
The DSRM-HRL framework addresses fairness in interactive recommendation systems by treating biased user interaction data...
Fairness Begins with State: Purifying Latent Preferences for Hierarchical Reinforcement Learning in Interactive Recommendation
The DSRM-HRL framework redefines fairness in interactive recommendation systems as a latent state purification problem r...
Zero-Knowledge Proof (ZKP) Authentication for Offline CBDC Payment System Using IoT Devices
A new model combines secure elements (SEs), zero-knowledge proofs (ZKPs), and intermittent synchronization to enable pri...
Zero-Knowledge Proof (ZKP) Authentication for Offline CBDC Payment System Using IoT Devices
A novel research model proposes a privacy-preserving offline CBDC payment system designed for resource-constrained IoT d...
Zero-Knowledge Proof (ZKP) Authentication for Offline CBDC Payment System Using IoT Devices
A novel research architecture proposes using Zero-Knowledge Proof (ZKP) authentication combined with Secure Elements (SE...
Zero-Knowledge Proof (ZKP) Authentication for Offline CBDC Payment System Using IoT Devices
A new research model enables secure offline Central Bank Digital Currency (CBDC) payments on Internet of Things devices ...
Zero-Knowledge Proof (ZKP) Authentication for Offline CBDC Payment System Using IoT Devices
A novel research model proposes a privacy-preserving offline Central Bank Digital Currency (CBDC) system designed specif...
DisenReason: Behavior Disentanglement and Latent Reasoning for Shared-Account Sequential Recommendation
DisenReason is a novel AI model that addresses shared-account sequential recommendation (SSR) by disentangling multiple ...
DisenReason: Behavior Disentanglement and Latent Reasoning for Shared-Account Sequential Recommendation
DisenReason is a novel two-stage AI model for shared-account sequential recommendation that fundamentally reframes the p...
DisenReason: Behavior Disentanglement and Latent Reasoning for Shared-Account Sequential Recommendation
DisenReason is a novel AI architecture for shared-account sequential recommendation that uses frequency-domain behavior ...
DisenReason: Behavior Disentanglement and Latent Reasoning for Shared-Account Sequential Recommendation
DisenReason is a novel AI architecture for Shared-account Sequential Recommendation (SSR) that dynamically infers the nu...
DisenReason: Behavior Disentanglement and Latent Reasoning for Shared-Account Sequential Recommendation
DisenReason is a novel AI method for shared-account sequential recommendation (SSR) that addresses the challenge of mult...
Not All Candidates are Created Equal: A Heterogeneity-Aware Approach to Pre-ranking in Recommender Systems
ByteDance researchers developed the Heterogeneity-Aware Adaptive Pre-ranking (HAP) framework to address gradient conflic...
Not All Candidates are Created Equal: A Heterogeneity-Aware Approach to Pre-ranking in Recommender Systems
ByteDance researchers introduced Heterogeneity-Aware Adaptive Pre-ranking (HAP), a novel framework addressing gradient c...
Not All Candidates are Created Equal: A Heterogeneity-Aware Approach to Pre-ranking in Recommender Systems
ByteDance researchers developed the Heterogeneity-Aware Adaptive Pre-ranking (HAP) framework to address gradient conflic...
Not All Candidates are Created Equal: A Heterogeneity-Aware Approach to Pre-ranking in Recommender Systems
ByteDance researchers developed the Heterogeneity-Aware Adaptive Pre-ranking (HAP) framework to address gradient conflic...
Not All Candidates are Created Equal: A Heterogeneity-Aware Approach to Pre-ranking in Recommender Systems
ByteDance researchers developed the Heterogeneity-Aware Adaptive Pre-ranking (HAP) framework to address gradient conflic...
IntroductionDMD-augmented Unpaired Neural Schr\"odinger Bridge for Ultra-Low Field MRI Enhancement
Researchers developed an AI framework that enhances ultra-low-field 64 millitesla (mT) brain MRI scans to resemble high-...