SE-RRM: 2M-Parameter Equivariant Model Beats HRM/TRM on Sudoku and ARC-AGI

gklambauer · x · 2026-09-05

Responding to a thread on Flow Reasoning Models, the authors highlight their ICML 2026 paper Symbol-Equivariant Recurrent Reasoning Models (SE-RRMs), extending the HRM/TRM family of tiny recursive reasoners.

Code is open-sourced. The quoted Flow Reasoning Models work applies continuous flows to discrete data, recurrently refining mistakes via self-conditioning on structured problems like Sudoku.

Original post →

More from Research

Research channel →