Entropy Meets Importance: A Unified Head Importance-Entropy Score for Stable and Efficient Transformer Pruning
"模型剪枝" 相关文章
Entropy Meets Importance: A Unified Head Importance-Entropy Score for Stable and Efficient Transformer Pruning
PermLLM: Learnable Channel Permutation for N:M Sparse Large Language Models
BaldWhisper: Faster Whisper with Head Shearing and Layer Merging
Fewer Weights, More Problems: A Practical Attack on LLM Pruning
OptiFLIDS: Optimized Federated Learning for Energy-Efficient Intrusion Detection in IoT
OptiFLIDS: Optimized Federated Learning for Energy-Efficient Intrusion Detection in IoT
Pruning and Distilling LLMs Using NVIDIA TensorRT Model Optimizer
My Notes From Spark+AI Summit 2020 (Application-Agnostic Talks)
PATCH: Learnable Tile-level Hybrid Sparsity for LLMs
Walk and Read Less: Improving the Efficiency of Vision-and-Language Navigation via Tuning-Free Multimodal Token Pruning
北京内推 | AMD北京AI算法团队招聘模型量化/剪枝算法实习生（可远程）
EvoP: Robust LLM Inference via Evolutionary Pruning
ICML 2025 | CoTo：让LoRA训练「渐入佳境」，模型融合、剪枝样样精通
ICML 2025 | CoTo：让LoRA训练「渐入佳境」，模型融合、剪枝样样精通
ICLR 2025 | 高效又稳定！人大团队提出模型剪枝新方法LLM-Streamline
英伟达联手MIT清北发布SANA 1.5，线性扩散Transformer再刷文生图新SOTA
AutoSculpt: A Pattern-based Automated Pruning Framework Designed to Enhance Efficiency and Accuracy by Leveraging Graph Learning and Deep Reinforcement Learning
12%计算量就能媲美原模型，Adobe、罗切斯特大学等提出YOPO剪枝技术
「TVM 教程」在 CPU 上部署 Hugging Face 剪枝模型
