文章预览
LG - 机器学习 CV - 计算机视觉 CL - 计算与语言 1、[LG] Learning to (Learn at Test Time)：RNNs with Expressive Hidden States 2、[LG] Mixture of A Million Experts 3、[CL] Collaborative Quest Completion with LLM-driven Non-Player Characters in Minecraft 4、[LG] On scalable oversight with weak LLMs judging strong LLMs 5、[LG] An Adaptive Stochastic Gradient Method with Non-negative Gauss-Newton Stepsizes 摘要：边测试边学习、百万专家混合模型、在Minecraft中与LLM驱动的非玩家角色协作完成任务、关于使用弱LLM判断强LLM的可扩展监管、采用非负高斯-牛顿步长的自适应随机梯度法 1、[LG] Learning to (Learn at Test Time): RNNs with Expressive Hidden States Y Sun, X Li, K Dalal, J Xu… [Stanford University & UC San Diego & UC Berkeley] 边测试边学习：具有可表达隐藏状态的RNN 要点: 提出TTT(测试时训练)层，一种新的序列建模层，其隐状态是一个模型，更新规则是自监督
………………………………
