finding
active
finding:geometry-behavior-correlate-robust-to-pooling-strategy-distance-metric-and-frozen-encoderGeometry-behavior correlate robust to pooling strategy, distance metric, and frozen encoder
Robustness checks confirm sign stability.
Source paper
extracted_from(2025) · Edward Yi Chang · Kaya, Zeyneb N. · Ethan Chang
Neighborhood — ranked by edge-count
Communities (2)
community
- Explores geometry of activation/behavior manifolds to enable selective, non-destructive concept interventions.
- Using second-order statistics to compress activation patterns while preserving feature co-occurrence structure, tested on genomic prediction tasks without large labeled datasets.
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.
- Relationship where geometry summaries predict behavioral thresholds.
- Core finding: the structure models use internally (representations) is precisely reflected in their external behavior (outputs).
- Author’s interpretive claim that the shared geometry is general and robust.
- Key geometry-to-behavior bridge finding in E3; robust to pooling choice, cosine vs. L2, and frozen external encoder
- The paper's generalization claim, asserting that the days-of-week finding scales to other cyclic and structured concepts.
- The bidirectional correspondence M_h ↔ M_y, indicating that geometry in representation is not incidental but causally shapes behavior.
- The paper's finding that the alignment holds in both directions — from representation to behavior and from behavior back to representation space.