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] Using Deep Autoregressive Models as Causal Inference Engines 2、[RO] CurricuLLM:Automatic Task Curricula Design for Learning Complex Robot Skills using Large Language Models 3、[LG] State-free Reinforcement Learning 4、[LG] Decomposable Transformer Point Processes 5、[CL] Evaluation of OpenAI o1:Opportunities and Challenges of AGI 摘要:把深度自回归模型用作因果推理引擎、利用大型语言模型学习复杂机器人技能的自动任务课程设计、无状态强化学习、可分解Transformer点过程、OpenAI o1评估 1、[LG] Using Deep Autoregressive Models as Causal Inference Engines D J Im, K Zhang, N Verma, K Cho [New York University & Columbia University] 把深度自回归模型用作因果推理引擎 要点: 提出了使用自回归语言模型作为因果推断引擎,适用于复杂的混杂变量和顺序动作。通过序列化数据实
………………………………