Attending to Memory Writes Reduces Seed Variability in Recurrent Length Generalization
cs.LG
Submitted: 2026-11-03
Updated: 2026-11-03
Code: https://github.com/swgoo/trace_relay
Terminology
Sources
- Delayed Attention Training Improves Length Generalization in Transformer--RNN Hybrids
- Short window attention enables long-term memorization
- Neural Machine Translation by Jointly Learning to Align and Translate
- Understanding and Improving Length Generalization in Recurrent Models
- Pretraining Recurrent Networks without Recurrence
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis