FACTS Grounding:用于评估大型语言模型(LLMs)基于长文档生成事实性(factuality )准确回答能力的基准测试
幻觉(Hallucinations,即事实性不准确的回答)依然困扰着大型语言模型(LLMs)。尤其是在面对更复...
Read MorePCNDE框架(Physics-constrained coupled neural differential equations,物理约束神经微分方程):实现较高精度和计算效率的1D血流模型
论文Physics-constrained coupled neural differential equat...
Read MoreTabPFN(Tabular Prior-data Fitted Network),这是一种基于生成式Transformer模型的表格数据基础模型
论文Accurate predictions on small data with a tabular fou...
Read More浅谈In-Context Learning (ICL),“上下文学习” 、 “情境学习”或者“内在学习”
In-Context Learning (ICL) 可翻译为 “上下文学习” 、 “情境学习”或者“内在学习”...
Read MoreGraphToken:实现图推理(graph reasoning)能力显著提升
论文Let Your Graph Do the Talking: Encoding Structured Da...
Read MoreNVIDIA推出Cosmos世界基础模型(World Foundation Model)平台
NVIDIA 宣布推出 NVIDIA Cosmos™,一个包含最先进的生成性世界基础模型、先进的标记器、保护机...
Read More基于大型语言模型(LLM)的垂直人工智能(Vertical AI)代理系统
论文Agentic Systems: A Guide to Transforming Industries w...
Read More奥特曼(Sam Altman)的新年Reflections(思考/反思)
OpenAI首席执行官山姆·奥特曼(Sam Altman)刚发了篇博文,标题为Reflections(思考/反...
Read More适宜性(appropriateness)理论及其在生成式人工智能中的应用
论文A theory of appropriateness with applications to gene...
Read More利用基础模型(Foundation Models)实现人工生命(Artificial Life)模拟的自动化搜索
论文Automating the Search for Artificial Life with Founda...
Read More采用结合深度访谈和生成式人工智能的代理构建方法,模拟1,052名真实个体
论文Generative Agent Simulations of 1,000 People提出了一种基于生成...
Read More自监督学习(Self-Supervised Learning,SSL)在单细胞基因组学(Single-Cell Genomics,SCG)中的有效应用
论文Delineating the effective use of self-supervised lear...
Read More扩散模型(diffusion model)的泛化源于几何自适应谐波表示(geometry-adaptive harmonic representation)
深度神经网络(Deep Neural Networks, DNNs)通过扩散模型表现出在图像生成和去噪任务中的...
Read More






