model architecture

Attribution Patching Outperforms Automated Circuit Discovery EMNLP 2024 Aux-NAS: Exploiting Auxiliary Labels with Negligibly Extra Inference Cost ICLR 2024 B-cosification: Transforming Deep Neural Networks to be Inherently Interpretable NeurIPS 2024 Beyond Performance Plateaus: A Comprehensive Study on Scalability in Speech Enhancement Interspeech 2024 Beyond the Limits: A Survey of Techniques to Extend the Context Length in Large Language Models IJCAI 2024 Boosting Vanilla Lightweight Vision Transformers via Re-parameterization ICLR 2024 Breaking the Curse of Multilinguality with Cross-lingual Expert Language Models EMNLP 2024 Building Optimal Neural Architectures using Interpretable Knowledge CVPR 2024 Chunk, Align, Select: A Simple Long-sequence Processing Method for Transformers ACL 2024 Conformer without Convolutions Interspeech 2024 Converting Transformers to Polynomial Form for Secure Inference Over Homomorphic Encryption ICML 2024 Data Driven Threshold and Potential Initialization for Spiking Neural Networks AISTATS 2024 Deep Neural Networks via Complex Network Theory: A Perspective IJCAI 2024 DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models ACL 2024 EAN: An Efficient Attention Module Guided by Normalization for Deep Neural Networks AAAI 2024 Efficient and Interpretable Grammatical Error Correction with Mixture of Experts EMNLP 2024 Fixed Non-negative Orthogonal Classifier: Inducing Zero-mean Neural Collapse with Feature Dimension Separation ICLR 2024 GradTree: Learning Axis-Aligned Decision Trees with Gradient Descent AAAI 2024 Grafting Vision Transformers WACV 2024 Graph Neural Networks Need Cluster-Normalize-Activate Modules NeurIPS 2024 GreedyViG: Dynamic Axial Graph Construction for Efficient Vision GNNs CVPR 2024 Harder Task Needs More Experts: Dynamic Routing in MoE Models ACL 2024 How does Architecture Influence the Base Capabilities of Pre-trained Language Models? A Case Study Based on FFN-Wider and MoE Transformers NeurIPS 2024 HybridBERT - Making BERT Pretraining More Efficient Through Hybrid Mixture of Attention Mechanisms NAACL 2024 HydraViT: Stacking Heads for a Scalable ViT NeurIPS 2024