Geometric Irreversibility and Dissipation Bounds in Non-Reciprocal Neural Attractor Dynamics
This paper demonstrates that in non-reciprocal neural attractor networks, a solenoidal drift component decouples equilibrium-like occupancies from irreversible dynamics, establishing a quantitative triad where path asymmetry, entropy production, and thermodynamic precision are governed by non-reciprocal currents despite an unchanged scalar landscape.