How Lossless Is Lossless Speculative Decoding? The Role of Numerical Precision in Orthrus Paper • 2609.15504 • Published 27 days ago • 37
CRISP: Cliff-awaRe Input-adaptive Sparse Prefilling with Structural-Mass-Motivated Routing Paper • 2609.01925 • Published Sep 1 • 7
Orthrus: Memory-Efficient Parallel Token Generation via Dual-View Diffusion Paper • 2605.12825 • Published May 12 • 13
SNOOPI: Supercharged One-step Diffusion Distillation with Proper Guidance Paper • 2412.02687 • Published Dec 3, 2024 • 114
Taipan: Efficient and Expressive State Space Language Models with Selective Attention Paper • 2410.18572 • Published Oct 24, 2024 • 18
Taipan: Efficient and Expressive State Space Language Models with Selective Attention Paper • 2410.18572 • Published Oct 24, 2024 • 18