Fork the consciousness, or download the project and create your own.

Malu Zhang

University of Electronic Science and Technology of China

Builder The claim is about what the hardware does, and the philosophy is left open

Malu Zhang is a researcher at the University of Electronic Science and Technology of China in Chengdu, working on spiking neural networks and brain-inspired computation, with prior affiliations at the National University of Singapore and Northumbria University. An h-index of 27 over 137 works marks a sustained program in neuromorphic architectures rather than a single result.

The program’s direction is to take a circuit mechanism measured in the brain and rebuild it as a machine-learning component. In “SpikeMoE: Brain-Inspired Competitive Routing for Flexible Spiking Mixture-of-Experts” (arXiv:2610.01418, posted October 1, 2026, with Xiaoli Liu, Yujie Liang, Jialin Li, and Chenxiao Dou), the copied mechanism is the k-winner-take-all competition of hippocampal CA1, implemented as lateral inhibition plus a refractory period in a spike-based router that selects Top-K experts by discrete spike counts.

For consciousness science the work sits on the substrate side of the question. Competitive inhibitory routing is one of the circuit motifs consciousness theories point to when they ask how a system selects what enters a workspace. Demonstrating that the same motif runs efficiently in silicon, and carries multimodal workloads under missing inputs, is evidence about what that motif buys functionally, which is the empirical grounding any account of emerging consciousness in non-biological substrates eventually needs.

Known for. Spiking neural networks, brain-inspired routing, spiking mixture-of-experts architectures

Coverage on this site

Related researchers