Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Why Post-Norm Transformers Collapse: Attention Amplification and Gradient Repair Failure
Xingjian Wang, Qingyu Han, Xiaodong Luo +1
Deep decoder-only Transformers often replace the original Post-Norm architecture with Pre-Norm variants because Post-Norm training is highly sensitive to warmup and learning rate u…
cs.LG2025
Feature Augmentation of GNNs for ILPs: Local Uniqueness Suffices
Qingyu Han, Qian Li, Linxin Yang +3
Integer Linear Programs (ILPs) are central to real-world optimizations but notoriously difficult to solve. Learning to Optimize (L2O) has emerged as a promising paradigm, with Grap…