Universal interpolation for deep residual self-attention networks
cs.LG
Submitted: 2026-10-01
Updated: 2026-10-01
Code: https://github.com/sibyllema/attention_universality
Sources
- Achieving Universal Approximation and Universal Interpolation via Nonlinearity of Control Families
- Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach
- Nanbeige4.2-3B: Unlocking Agentic Capabilities in a Compact Model
- Universal Approximation Power of Deep Residual Neural Networks via Nonlinear Control Theory
- Hyperloop Transformers
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