Neural Network Optimization

Loss Landscape of Shallow ReLU-like Neural Networks: Stationary Points, Saddle Escape, and Network Embedding ICLR 2025 LSSF: Safety Alignment for Large Language Models through Low-Rank Safety Subspace Fusion ACL 2025 Mamba4D: Efficient 4D Point Cloud Video Understanding with Disentangled Spatial-Temporal State Space Models CVPR 2025 Mamba-Adaptor: State Space Model Adaptor for Visual Recognition CVPR 2025 Mamba YOLO: A Simple Baseline for Object Detection with State Space Model AAAI 2025 Marrying LLMs with Dynamic Forecasting: A Graph Mixture-of-expert Perspective NAACL 2025 MARS: Unleashing the Power of Variance Reduction for Training Large Models ICML 2025 Masks Can be Learned as an Alternative to Experts ACL 2025 Mean-field analysis of polynomial-width two-layer neural network beyond finite time horizon COLT 2025 Memory-Efficient Backpropagation for Fine-Tuning LLMs on Resource-Constrained Mobile Devices EMNLP 2025 Memory or Reasoning? Explore How LLMs Compute Mixed Arithmetic Expressions ACL 2025 MEPT: Mixture of Expert Prompt Tuning as a Manifold Mapper EMNLP 2025 MERIT: Maximum-normalized Element-wise Ratio for Language Model Large-batch Training ICML 2025 MGCFNN: A Neural MultiGrid Solver with Novel Fourier Neural Network for High Wave Number Helmholtz Equations ICLR 2025 Mitigating Attention Localization in Small Scale: Self-Attention Refinement via One-step Belief Propagation EMNLP 2025 Mitigating Confounding in Speech-Based Dementia Detection through Weight Masking ACL 2025 Mitigating Forgetting in Continual Learning with Selective Gradient Projection IJCNLP 2025 Mitigating Hallucinations in Large Vision-Language Models via DPO: On-Policy Data Hold the Key CVPR 2025 Mitigating Object Hallucinations in MLLMs via Multi-Frequency Perturbations EMNLP 2025 Mitigating Posterior Salience Attenuation in Long-Context LLMs with Positional Contrastive Decoding ACL 2025 Mitigating Selection Bias with Node Pruning and Auxiliary Options ACL 2025 Mitigating Visual Forgetting via Take-along Visual Conditioning for Multi-modal Long CoT Reasoning ACL 2025 Mix-LN: Unleashing the Power of Deeper Layers by Combining Pre-LN and Post-LN ICLR 2025 Mixture-of-Clustered-Experts: Advancing Expert Specialization and Generalization in Instruction Tuning EMNLP 2025 MLAAN: Scaling Supervised Local Learning with Multilaminar Leap Augmented Auxiliary Network AAAI 2025