long-range dependencies
A phenomenon in sequences where earlier parts of the data significantly influence later parts. In AI, it poses challenges for modeling and requires sophisticated approaches, especially in recurrent networks.
- A Closer Look at Graph Transformers: Cross-Aggregation and Beyond
- A Multimodal BiMamba Network with Test-Time Adaptation for Emotion Recognition Based on Physiological Signals
- BeyondMix: Leveraging Structural Priors and Long-Range Dependencies for Domain-Invariant LiDAR Segmentation
- Block-Biased Mamba for Long-Range Sequence Processing
- Context-Aware Regularization with Markovian Integration for Attention-Based Nucleotide Analysis
- Copresheaf Topological Neural Networks: A Generalized Deep Learning Framework
- DuSA: Fast and Accurate Dual-Stage Sparse Attention Mechanism Accelerating Both Training and Inference
- DualEqui: A Dual-Space Hierarchical Equivariant Network for Large Biomolecules
- EventMG: Efficient Multilevel Mamba-Graph Learning for Spatiotemporal Event Representation
- Fractional Diffusion Bridge Models
- Generative Graph Pattern Machine
- Hierarchical Information Aggregation for Incomplete Multimodal Alzheimer's Disease Diagnosis
- JanusDNA: A Powerful Bi-directional Hybrid DNA Foundation Model
- L-MTP: Leap Multi-Token Prediction Beyond Adjacent Context for Large Language Models
- Learning Sparse Approximate Inverse Preconditioners for Conjugate Gradient Solvers on GPUs
- MODEM: A Morton-Order Degradation Estimation Mechanism for Adverse Weather Image Recovery
- MVSMamba: Multi-View Stereo with State Space Model
- Overcoming Long Context Limitations of State Space Models via Context Dependent Sparse Attention
- PPMStereo: Pick-and-Play Memory Construction for Consistent Dynamic Stereo Matching
- Predicting partially observable dynamical systems via diffusion models with a multiscale inference scheme
- RHYTHM: Reasoning with Hierarchical Temporal Tokenization for Human Mobility
- Return of ChebNet: Understanding and Improving an Overlooked GNN on Long Range Tasks
- SALoM: Structure Aware Temporal Graph Networks with Long-Short Memory Updater
- Safe-Sora: Safe Text-to-Video Generation via Graphical Watermarking
- Spectral Convolutional Conditional Neural Process
- Towards Multiscale Graph-based Protein Learning with Geometric Secondary Structural Motifs
- VORTA: Efficient Video Diffusion via Routing Sparse Attention
- WaLRUS: Wavelets for Long range Representation Using State Space Methods