Mean Pooling

Standard baseline aggregation method that covariance pooling improves upon; discards joint activation structure.

Neighborhood — ranked by edge-count

concept

Sequence pooling
extends
General technique for aggregating variable-length token embedding sequences into fixed-length representations; mean pooling and covariance pooling are instances.
First moments
implements
Statistical moments capturing per-feature central tendency; the paper argues first-moment pooling (mean pooling) destroys joint activation structure.

claim

Second moments preserve structure that first moments destroy.
contradicts
Core interpretive claim generalizing beyond genomics; argues mean pooling discards information present in covariance.

method

Covariance Pooling
extends
Novel aggregation technique replacing mean pooling; preserves joint activation structure (feature co-occurrence) in token embeddings.

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

biddingconcept0.737
The act of placing monetary bids in the auction phase.
tunnelingconcept0.722
Quantum-physics-inspired notion of a direct connection between matter and the I-plenum, allowing centers to reveal the I.
Chunkingconcept0.721
Rescaling of search to a higher organizational level; hypothesised as intrinsic to ETIs.
Slidingmethod0.716
Dynamic condition: smooth movement of text across the screen.
Subdivision of lotsconcept0.715
The process of splitting larger lots into smaller ones to increase density while maintaining individual ownership.
commitmentconcept0.712
An internal obligation to make some sentence true, a key abstraction for Elephant speech acts.
probabilityconcept0.711
Measure of expected sensory input, core to linking value and surprise.
gene pool of sequencesconcept0.709
A shared repository (the Internet) where morphogenetic sequences are stored, exchanged, mutated, and improved by the public.