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、[RO] ARCap：Collecting High-quality Human Demonstrations for Robot Learning with Augmented Reality Feedback 2、[LG] Unintentional Unalignment： Likelihood Displacement in Direct Preference Optimization 3、[LG] Loss Landscape Characterization of Neural Networks without Over-Parametrization 4、[LG] Multi-modal graph neural networks for localized off-grid weather forecasting 5、[LG] Dualformer：Controllable Fast and Slow Thinking by Learning with Randomized Reasoning Traces 摘要：利用增强现实反馈为机器人学习收集高质量的人工演示、直接偏好优化中的似然位移、无过参数化神经网络的损失景观刻画、面向局部非格点位置天气预报的多模态图神经网络、通过随机推理轨迹学习实现可控的快慢思考 1、[RO] ARCap: Collecting High-quality Human Demonstrations for Robot Learning with Augmented Reality Feedback S Ch
………………………………
