On the Residual Scaling of Looped Transformers: Stability and Transferability
cs.LG
Submitted: 2026-06-16
Updated: 2026-09-11
Comments: 19 pages, 9 figures
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- The Llama 3 Herd of Models
- Don't be lazy: CompleteP enables compute-efficient deep transformers
- LLaMA: Open and Efficient Foundation Language Models
- Completed Hyperparameter Transfer across Modules, Width, Depth, Batch and Duration
- Loop Neural Networks for Parameter Sharing
- Understanding Warmup-Stable-Decay Learning Rates: A River Valley Loss Landscape Perspective
- Tensor Programs V: Tuning Large Neural Networks via Zero-Shot Hyperparameter Transfer
- Scaling Latent Reasoning via Looped Language Models
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