1
Unifying Physical Backpropagation
统一物理反向传播框架，连接神经科学与物理计算，为硬件训练算法提供理论新视角。
arXiv:2608.11585v1 Announce Type: cross Abstract: Physical computing systems exploit device dynamics for computation, but their gradient-based optimiz…
2
Thermodynamic Irreversibility of Training Algorithms
从热力学视角揭示训练算法的不可逆本质，为理解深度学习优化过程提供全新理论框架。
arXiv:2605.21933v1 Announce Type: cross Abstract: The training algorithms for AI systems all introduce far-from-equilibrium dynamical processes, and u…
3
Good flavor search in SU(5): a machine learning approach
用机器学习搜索SU(5)风味模型，物理与AI交叉的前沿探索
arXiv:2511.08154v2 Announce Type: replace-cross Abstract: We revisit the fermion mass problem of the $SU(5)$ grand unified theory using machine learni…