Batch-Adaptive Path Selection for Efficient Linear Attention Prefill

PSPO: Trainable Potential-Based Reward Shaping with Internal Model Signals for Post-Training Policy Optimization of Large Language Models

Energy Measurement Study for Vision-Language Model Inference: A Single-GPU Empirical Analysis

EDG-ODE: Edge-Driven High-Order Graph Neural ODE for Continuous Traffic Flow Forecasting

Semi-Supervised Mixture Models under the Concept of Missing at Random with Margin Confidence and Aranda-Ordaz Function

An Adaptive Federated Learning Client Selection Method Based on Distillation Calibration

A Test Script Generation Method Based on Large Language Models with Model Sensitivity Analysis in Semi-Physical Simulation

Prompt-Guided Hierarchical Metric Learning with Cross-Level Similarity Consistency

A Gaussian Distribution-Based Oversampling Method On SPD Manifolds For Imbalanced Physiological Signal Classification

FedSPADE: A Novel Federated Learning Method with Skew-Aware Personalized Aggregation and Distillation Enhancement

Offline Clustering and Online Incentive Mechanism for Personalized Collaborative Machine Learning

Group Enhanced Knowledge Tracing Model

LRHR-Net: Coarse-to-Fine Flow Field Reconstruction for Scramjet Combustor with A Wide Speed Range

A semi-supervised online sequential framework for complex data stream classification

DynaMed: A Dynamic Causal Reasoning Framework for Medication Recommendation

Label Distribution Learning with Label-cluster-specific Features