爱可可 AI 前沿推介(6.4)

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 RO - 机器人 1、[LG] Transformers are SSMs：Generalized Models and Efficient Algorithms Through Structured State Space Duality 2、[LG] Fully Unconstrained Online Learning 3、[LG] Scaling Laws for the Value of Individual Data Points in Machine Learning 4、[CL] Perplexed by Perplexity：Perplexity-Based Data Pruning With Small Reference Models 5、[CV] Video-MME：The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis 摘要：通过结构化状态空间对偶性实现通用化模型和高效算法、完全无约束在线学习、机器学习单数据点价值的缩放率、用小型参考模型进行基于困惑度的数据裁剪、视频分析多模态LLM的综合评估基准 1、[LG] Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality T Dao, A Gu [Princeton University & CMU] Transformer与SSM的统一性：通 ………………………………