Zhen Qin

7

Papers

32

Total Citations

Papers (7)

CO2: Efficient Distributed Training with Full Communication-Computation Overlap

Exploring Transformer Extrapolation

Hybrid Latent Reasoning via Reinforcement Learning

Various Lengths, Constant Speed: Efficient Language Modeling with Lightning Attention

Resisting Backdoor Attacks in Federated Learning via Bidirectional Elections and Individual Perspective

Deep Non-Rigid Structure-from-Motion Revisited: Canonicalization and Sequence Modeling

Federated Full-Parameter Tuning of Billion-Sized Language Models with Communication Cost under 18 Kilobytes