Rethinking Cross-Layer Information Routing in Diffusion Transformers Paper • 2605.20708 • Published 7 days ago • 98
Full Attention Strikes Back: Transferring Full Attention into Sparse within Hundred Training Steps Paper • 2605.16928 • Published 11 days ago • 90
EdgeRazor: A Lightweight Framework for Large Language Models via Mixed-Precision Quantization-Aware Distillation Paper • 2605.04062 • Published Apr 10 • 30