Flux Attention: Context-Aware Hybrid Attention for Efficient LLMs Inference Paper • 2604.07394 • Published 8 days ago • 16
Hard-aware Instance Adaptive Self-training for Unsupervised Cross-domain Semantic Segmentation Paper • 2302.06992 • Published Feb 14, 2023
Elastic Attention: Test-time Adaptive Sparsity Ratios for Efficient Transformers Paper • 2601.17367 • Published Jan 24 • 34