finding
active
finding:hierarchical-architectures-outperform-others-by-30-40-on-scr-and-absorptionHierarchical architectures outperform others by 30-40% on SCR and absorption
Margin of advantage.
Source paper
extracted_from(2025) · Adam Karvonen · Can Rager · Johnny Lin · Curt Tigges +11
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Inverse SCR scaling persists across intervention sizes for non-hierarchical architecturesfinding0.792Rules out fraction-modified explanation.
- ReLU underperforms.
- Inverse scaling on SCR.
- Inverse scaling.
- Hierarchical NCA architectures could enhance convergence speed and stability in DiffLogic CA for complex shapeshypothesis0.738Future direction hypothesis for addressing optimization challenges in complex pattern generation
- Demonstrated CNN representations predict neurons in visual cortex; background motivation for neural-network-brain correspondence.
- Model-scale dependence.