Neural network learns inverse Robinson-Schensted correspondence via structured 2D token embeddings. 100% exact match on n=10.
-
Updated
Jul 5, 2026 - Python
Neural network learns inverse Robinson-Schensted correspondence via structured 2D token embeddings. 100% exact match on n=10.
Mechanistic interpretability of small transformers: RSK correspondence and Pythia-70m
Web pages for FPSAC
RIME: spectral, transport, accessibility, and observable-pipeline geometry for finite represented systems, with Rubik as a reproducible laboratory.
Counting k-bounded functions on [n]
My mathematics research page
Add a description, image, and links to the algebraic-combinatorics topic page so that developers can more easily learn about it.
To associate your repository with the algebraic-combinatorics topic, visit your repo's landing page and select "manage topics."