爱可可 AI 前沿推介(5.17)

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] Challenges in Deploying Long-Context Transformers：A Theoretical Peak Performance Analysis 2、[LG] ALPINE：Unveiling the Planning Capability of Autoregressive Learning in Language Models 3、[CL] Word2World：Generating Stories and Worlds through Large Language Models 4、[AI] People cannot distinguish GPT-4 from a human in a Turing test 5、[AS] Naturalistic Music Decoding from EEG Data via Latent Diffusion Models 摘要：部署长上下文Transformer的挑战、揭示语言模型自回归学习的规划能力、用大语言模型生成故事和世界、在图灵测试中人无法区分GPT-4和人类、基于潜扩散模型的脑电数据自然音乐解码 1、[LG] Challenges in Deploying Long-Context Transformers: A Theoretical Peak Performance Analysis Y Fu [University of Edinburgh] 部署长上下文Transformer的挑战：理论峰值性能分析 要点: 与短上下文(例 ………………………………