EnergyCal: Taming Loss Spikes in Muon-Family Optimizers for Sparse MoE Training Paper • 2610.37062 • Published 9 days ago
MESH: Memory-Efficient Sinkhorn Optimization for Mixture-of-Experts Training Paper • 2608.04407 • Published Aug 5
DTrOCR: Decoder-only Transformer for Optical Character Recognition Paper • 2308.15996 • Published Aug 30, 2023 • 5
LayoutLLM: Large Language Model Instruction Tuning for Visually Rich Document Understanding Paper • 2403.14252 • Published Mar 21, 2024
DTrOCR: Decoder-only Transformer for Optical Character Recognition Paper • 2308.15996 • Published Aug 30, 2023 • 5