Ember: A Kilobyte-State Optimizer for Token Interfaces

arXiv:2607.01455 · cs.LG, cs.AI · Submitted 2026-07-01 · Read on arXiv

cs.LG, cs.AI

Submitted: 2026-07-01

Updated: 2026-10-11

Terminology

• Adam — 1× in this paper · explained in 2 episodes
• Transformer — 1× in this paper · explained in 2 episodes

Sources

• Modular Duality in Deep Learning
• DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
• Symbolic Discovery of Optimization Algorithms
• Scalable Second Order Optimization for Deep Learning
• Proximal Policy Optimization Algorithms
• Symmetry-Compatible Principle for Optimizer Design: Embeddings, LM Heads, SwiGLU MLPs, and MoE Routers
• Gradient Descent Happens in a Tiny Subspace

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks