Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Enhancing SignSGD: Small-Batch Convergence Analysis and a Hybrid Switching Strategy
Haoran Chen, Wentao Wang
SignSGD compresses each stochastic gradient coordinate to a single bit, offering substantial memory and communication savings, but its 1-bit quantization removes magnitude informat…
cs.LG2025
Revisit Modality Imbalance at the Decision Layer
Xiaoyu Ma, Hao Chen
Multimodal learning integrates information from different modalities to enhance model performance, yet it often suffers from modality imbalance, where dominant modalities overshado…