MedAlign: A Synergistic Framework of Multimodal Preference Optimization and Federated Meta-Cognitive Reasoning
"混合专家架构" 相关文章
MTmixAtt: Integrating Mixture-of-Experts with Multi-Mix Attention for Large-Scale Recommendation
Mix- and MoE-DPO: A Variational Inference Approach to Direct Preference Optimization
Character.AI Open Sources pipeling-sft: A Scalable Framework for Fine-Tuning MoE LLMs like DeepSeek V3
美团开源推理大模型 LongCat-Flash-Thinking
Dynamic Adaptive Parsing of Temporal and Cross-Variable Patterns for Network State Classification
MiniMax-M1：闪电注意力重塑大模型推理效率，百万上下文时代来临，附技术报告英中对照版
被 OpenAI 点名后，智谱刚刚发布新开源模型：国产最全能，一句话造出搜索引擎
Mixture of Experts in Large Language Models
苦等一年 Meta终于放大招 正式发布开源大模型Llama 4
Meta 发布 Llama 4 系列 AI 模型，引入“混合专家架构”提升效率
Meta releases Llama 4, a new crop of flagship AI models
华尔街深度研究：DeepSeek是AI末日吗？
Hunyuan-Large and the MoE Revolution: How AI Models Are Growing Smarter and Faster
全球首次！时序大模型突破十亿参数，华人团队发布Time-MoE，预训练数据达3000亿个时间点
全球首次，时序大模型突破十亿参数，华人团队发布Time-MoE，预训练数据达3000亿个时间点
微軟釋出3款Phi-3.5模型
