pith. sign in

Exploring activation pat- terns of parameters in language models

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.LG 1

years

2026 1

verdicts

UNVERDICTED 1

clear filters

representative citing papers

Graph Memory Transformer (GMT)

cs.LG · 2026-04-26 · unverdicted · novelty 7.0

Graph Memory Transformer replaces FFN sublayers with a graph memory cell using 128 centroids and transition matrices per block, yielding stable training at 82.2M parameters but higher validation loss than a 103M dense baseline.

citing papers explorer

Showing 0 of 0 citing papers after filters.

No citing papers match the current filters.