A Technique of Pure Reason
"模型效率" 相关文章
This AI paper from DeepSeek-AI Explores How DeepSeek-V3 Delivers High-Performance Language Modeling by Minimizing Hardware Overhead and Maximizing Computational Efficiency
斯坦福 AI 指数报告发布：2024 年美国 AI 私人投资约为中国的 12 倍、GPT-3.5级别模型推理成本下降 280 倍
Decoupling Tokenization: How Over-Tokenized Transformers Redefine Vocabulary Scaling in Language Models
Autonomy-of-Experts (AoE): A Router-Free Paradigm for Efficient and Adaptive Mixture-of-Experts Models
面壁、智谱，卷上新赛道
Apple Researchers Introduce Instruction-Following Pruning (IFPruning): A Dynamic AI Approach to Efficient and Scalable LLM Optimization
田渊栋：2024年年终总结
1-bit大模型还能再突破！新一代BitNet架构启用4位激活值
【NLP】Kaggle知识点：文本分类与LoRA
清华UCSD提出全新微调方法，8B小模型媲美GPT-4o！科学问题正确率提高28%
Google AI Introduces LAuReL (Learned Augmented Residual Layer): Revolutionizing Neural Networks with Enhanced Residual Connections for Efficient Model Performance
Heterogeneous Mixture of Experts (HMoE): Enhancing Model Efficiency and Performance with Diverse Expert Capacities
成本降低90%！Claude上新Gemini同款提示词缓存，一次性记住整个代码库
This AI Paper by Meta FAIR Introduces MoMa: A Modality-Aware Mixture-of-Experts Architecture for Efficient Multimodal Pre-training
