The Hidden Power of Normalization: Exponential Capacity Control in Deep Neural Networks
"泛化能力" 相关文章
Leveraging Hierarchical Image-Text Misalignment for Universal Fake Image Detection
A Retrospect to Multi-prompt Learning across Vision and Language
Mutual Information guided Visual Contrastive Learning
Ariadne: A Controllable Framework for Probing and Extending VLM Reasoning Boundaries
Exploring Landscapes for Better Minima along Valleys
Metis-SPECS: Decoupling Multimodal Learning via Self-distilled Preference-based Cold Start
MGA: Memory-Driven GUI Agent for Observation-Centric Interaction
世界模型==VQA？机器人不用想象画面，预测语义就够了
Dynamic Dropout: Leveraging Conway's Game of Life for Neural Networks Regularization
Training data membership inference via Gaussian process meta-modeling: a post-hoc analysis approach
Xihe: Scalable Zero-Shot Time Series Learner Via Hierarchical Interleaved Block Attention
On Generalization in Agentic Tool Calling: CoreThink Agentic Reasoner and MAVEN Dataset
Reasoning Models Reason Well, Until They Don't
Predictive Coding Enhances Meta-RL To Achieve Interpretable Bayes-Optimal Belief Representation Under Partial Observability
用「进化+压力测试」自动生成的竞赛级编程题，各家大模型谁更hold住？
$\alpha$-LoRA: Effective Fine-Tuning via Base Model Rescaling
Making Classic GNNs Strong Baselines Across Varying Homophily: A Smoothness-Generalization Perspective
Making Classic GNNs Strong Baselines Across Varying Homophily: A Smoothness-Generalization Perspective
Metis-HOME: Hybrid Optimized Mixture-of-Experts for Multimodal Reasoning
