Kai Yi

Researcher@Meta·美国·OpenReview

研究方向

Large Language Models · Differential Privacy · Machine Learning Optimization · Federated Learning · Compression · Sparsity · Quantization · Transfer Learning · Data-Efficient Learning · Computer Vision

5.0

WiniQ: Accelerating Quantization-Aware Training of LLMs around Saddle Points

ICLR 2026Rejected

三作

4.0

Cohort Squeeze: Beyond a Single Communication Round per Cohort in Cross-Device Federated Learning

ICLR 2026Rejected

一作

3.5

Symmetric Pruning for Large Language Models

ICLR 2026Withdrawn

一作

4.2

Prune at the Clients, Not the Server: Accelerated Sparse Training in Federated Learning

ICLR 2025Withdrawn

二作

3.4

Cohort Squeeze: Beyond a Single Communication Round per Cohort in Cross-Device Federated Learning

ICLR 2025Withdrawn

一作

3.0

FedComLoc: Communication-Efficient Distributed Training of Sparse and Quantized Models

合作者 (20)