爱可可 AI 前沿推介(11.11)

文章预览

LG - 机器学习 CV - 计算机视觉 CL - 计算与语言 AS - 音频与语音 1、[CL] Mixture-of-Transformers：A Sparse and Scalable Architecture for Multi-Modal Foundation Models 2、[IR] Best Practices for Distilling Large Language Models into BERT for Web Search Ranking 3、[AS] Music Foundation Model as Generic Booster for Music Downstream Tasks 4、[LG] PatternBoost：Constructions in Mathematics with a Little Help from AI 5、[LG] Constrained Diffusion Implicit Models 摘要：多模态基础模型的稀疏可扩展架构、将大型语言模型蒸馏到BERT用于Web搜索排名的最佳实践、音乐基础模型作为音乐下游任务的通用助推器、人工智能辅助的数学结构发现、约束扩散隐式模型 1、[CL] Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models W Liang, L Yu, L Luo, S Iyer… [FAIR at Meta & Stanford University] Transformer混合：多模态基础模型的稀疏可扩展架构 要点: 混合Tr ………………………………