ScaleNet: Scaling up Pretrained Neural Networks with Incremental Parameters
"模型扩展" 相关文章
谷歌×耶鲁联手发布抗癌神器，AI推理精准狙击「隐身」癌细胞
Claude Skills are awesome, maybe a bigger deal than MCP
Scaling Laws and Symmetry, Evidence from Neural Force Fields
Continual Adapter Tuning with Semantic Shift Compensation for Class-Incremental Learning
QDeepGR4J: Quantile-based ensemble of deep learning and GR4J hybrid rainfall-runoff models for extreme flow prediction with uncertainty quantification
Understanding Generative Recommendation with Semantic IDs from a Model-scaling View
ICML 2025 | 隐空间记忆登场！M+打破上下文限制，8B模型记住160K+内容
Supernova: Achieving More with Less in Transformer Architectures
微软等提出「模型链」新范式，与Transformer性能相当，扩展性灵活性更好
不只靠“堆参数”：Qwen新突破ParScale，用“并行”让模型更聪明
Multimodal Models Don’t Need Late Fusion: Apple Researchers Show Early-Fusion Architectures are more Scalable, Efficient, and Modality-Agnostic
谷歌超硬核教科书来了！Jeff Dean带货揭Gemini训练秘籍：在TPU上scaling
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
Anthropic 联合创始人：AI 尚未发展至极限，2025 年将继续高速狂飙
Is AI progress slowing down?
Token化一切，甚至网络，北大&谷歌&马普所提出TokenFormer，Transformer从来没有这么灵活过
Token化一切，甚至网络！北大&谷歌&马普所提出TokenFormer，Transformer从来没有这么灵活过！
黄仁勋：我从不在乎市场份额，英伟达唯一目标是创造新市场
Microsoft Releases GRIN MoE: A Gradient-Informed Mixture of Experts MoE Model for Efficient and Scalable Deep Learning
