Error-Free Linear Attention is a Free Lunch: Exact Solution from Continuous-Time Dynamics
Paper • 2512.12602 • Published • 44
Golf
-
-
Diffusion Language Models are Super Data Learners
Paper • 2511.03276 • Published • 132 -
DoPE: Denoising Rotary Position Embedding
Paper • 2511.09146 • Published • 98 -
INT v.s. FP: A Comprehensive Study of Fine-Grained Low-bit Quantization Formats
Paper • 2510.25602 • Published • 80 -
XQuant: Achieving Ultra-Low Bit KV Cache Quantization with Cross-Layer Compression
Paper • 2510.11236 • Published • 1 -
Learnable Multipliers: Freeing the Scale of Language Model Matrix Layers
Paper • 2601.04890 • Published • 44 -
Quartet II: Accurate LLM Pre-Training in NVFP4 by Improved Unbiased Gradient Estimation
Paper • 2601.22813 • Published • 60