Skip to main content
Search

Shihezi University

2 related articles

Subscribe to updates
  1. Attention modules add nothing on nine-class chest X-rays: five add-ons tie an identical ResNet-50 across ten seeds

    On nine-class chest radiograph classification, five lightweight attention or graph modules were each inserted into the same ResNet-50 backbone, and none beat the unmodified baseline across ten random seeds. The baseline reached a per-seed macro-AUC of 0.7990±0.0023; CBAM and GAT were significantly lower after multiple-comparison correction, yet every effect size fell inside a pre-specified 0.01-AUC equivalence margin. The result held after a 2.4-fold larger training set, on ResNet-18, EfficientNet-B0 and MobileNetV2, and under a patient-level split.

  2. bioRxiv

    Kairos-GFM generates transition-state candidates in 0.0118 s via deterministic equivariant flow matching and recovers the target channel in 75.94% of 881 reactions

    The authors introduce Kairos-GFM, a deterministic equivariant flow-matching model that combines a two-dimensional reaction graph with the Kabsch-aligned mean of three-dimensional reactant and product geometries; on 881 Transition1x reactions whose reference transition states passed strict saddle-point and bidirectional intrinsic reaction coordinate validation, it generates one candidate in 0.0118 s with a median RMSD of 0.14 Å, and a hierarchical physical-validation framework reveals that low RMSD and saddle-point convergence do not guarantee the intended reaction channel, with the model recovering the target channel in 75.94% of reactions, outperforming diffusion-based baselines but remaining below transport-based methods.

You’re all caught up