Emergent One-Third Scaling Law as Attention Tries to Concentrate
cs.LG, cs.AI, stat.ML
Submitted: 2026-09-26
Updated: 2026-09-26
Terminology
Sources
- GPT-4 Technical Report
- A Theory for Emergence of Complex Skills in Language Models
- Chinchilla Scaling: A replication attempt
- Learning Curves for SGD on Structured Features
- A Dynamical Model of Neural Scaling Laws
- Theory of Scaling Laws for In-Context Regression: Depth, Width, Context and Time
- Neural Scaling Laws Rooted in the Data Distribution
- Deriving Neural Scaling Laws from the statistics of natural language
- Scaling Laws and Spectra of Shallow Neural Networks in the Feature Learning Regime
- The Pile: An 800GB Dataset of Diverse Text for Language Modeling
- Scaling Laws for Autoregressive Generative Modeling
- Deep Learning Scaling is Predictable, Empirically
- Training Compute-Optimal Large Language Models
- Learning Curve Theory
- Scaling Laws for Neural Language Models
- A Boundary-Layer Mechanism for One-Third Scaling in Online Softmax Classification
- Neural Scaling Universality: If Exponents Are Fixed, Time to Understand Coefficients
- Inverse Depth Scaling From Most Layers Being Similar
- Universal One-third Time Scaling in Learning Peaked Distributions
- Physics of Skill Learning
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: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks