William Brandon

~William_Brandon1

4

论文总数

2.0

年均投稿

6.2

平均评分

接收情况3/4

会议分布

NeurIPS

1

ICLR

1

ICML

1

COLM

1

发表论文 (4 篇)

20252 篇

Ladder Residual: Redefining Tensor Parallelism in Transformers for Accelerated Inference

ICLR 2025Rejected

Ladder-Residual: Parallelism-Aware Architecture for Accelerating Large Model Inference with Communication Overlapping

ICML 2025Poster

20242 篇

Reducing Transformer Key-Value Cache Size with Cross-Layer Attention

NeurIPS 2024Poster

Hydra: Sequentially-Dependent Draft Heads for Medusa Decoding

COLM 2024Poster

合作者 (14)

Jonathan Ragan-Kelley4 篇

Mayank Mishra3 篇

Ben Athiwaratkun2 篇

Muru Zhang2 篇

Shuaiwen Leon Song2 篇