Length Generalization Needs Proper Regularization

arXiv:2610.04518 · cs.LG · Submitted 2026-10-03 · Read on arXiv

cs.LG

Submitted: 2026-10-03

Updated: 2026-10-03

Code: https://github.com/deep-spin/len-gen-regularization

Terminology

• Generalization — 1× in this paper · explained in 13 episodes
• Regularization — 1× in this paper · explained in 3 episodes

Sources

• Layer Normalization
• Longformer: The Long-Document Transformer
• Generating Long Sequences with Sparse Transformers
• The Pile: An 800GB Dataset of Diverse Text for Language Modeling
• Mistral 7B
• Scalable-Softmax Is Superior for Attention
• NVIDIA Nemotron Nano 2: An Accurate and Efficient Hybrid Mamba-Transformer Reasoning Model
• Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
• Olmo 3
• BayesFormer: Transformer with Uncertainty Estimation
• Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
• LLaMA: Open and Efficient Foundation Language Models
• An Empirical Study of Mamba-based Language Models
• Qwen3 Technical Report

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