2 papers
cs.LG2026
LampAttention: Look-Ahead Mixed-Precision FlashAttention for Dedicated Accelerators
Stanislav Budzinskiy, Marian Gloser, Tolunay Yilmaz +5
While most attention logits can be computed in low precision without degrading numerical stability, current attention kernels fail to exploit this phenomenon. We introduce a novel…
cs.LG2026
LAMP: Look-Ahead Mixed-Precision Inference of Large Language Models
Stanislav Budzinskiy, Marian Gloser, Tolunay Yilmaz +5
Mixed-precision computations are a hallmark of the current stage of AI, driving the progress in large language models towards efficient, locally deployable solutions. This article…