2 papers
cs.LG2026
LionMuon: Alternating Spectral and Sign Descent for Efficient Training
Arman Bolatov, Artem Riabinin, Nikita Kornilov +6
In large-scale optimization, the cheapness and effectiveness of update steps are the most crucial factors for a successful optimizer. Sign-based optimizers like Lion or Signum prod…
cs.LG2026
Beyond Memorization: Extending Reasoning Depth with Recurrence, Memory and Test-Time Compute Scaling
Ivan Rodkin, Daniil Orel, Konstantin Smirnov +9
Reasoning is a core capability of large language models, yet how multi-step reasoning is learned and executed remains unclear. We study this question in a controlled cellular-autom…