4 papers
Reciprocity Separates Gradient Flow from Rotation in Conservative Physical Learning
Ruiwu Niu, Xiaowen Bi, Michaël Antonie van Wyk
Physical learning lets a trainable material or network use its own physical response to carry error signals, reducing the need for a separately programmed backward computation. We…
Message Order Selects Opposite Collective Opinions on Laplacian-Cospectral Networks
Ruiwu Niu, Xincheng Shu, Ying Zhao
Receiving the same messages in different orders can change an agent's judgement. We construct a network model in which this local sequence effect determines the final opinion of an…
Same Resident Strains, Different Attractors: Opposite Local Growth Signs for a Rare Third Strain
Ruiwu Niu, Xincheng Shu, Ying Zhao +1
Most models assess whether a newly introduced pathogen strain can grow when rare by testing it against a steady endemic background. Resident strains, however, can have more than on…
Reliability-Contagion Feasibility in LLM Multi-Agent Networks
Ruiwu Niu, Xincheng Shu, Ying Zhao
Communication allows large language model agents to pool evidence, but it also creates paths along which an erroneous claim can spread. We formulate a correction-aware network mode…