arXiv2
Search
Dark
/ Date
/ Name
Aa
W
/ Date
/ Name
"au:"Dazhao Cheng"" — arXiv2 Search
Showing 1–7 of 7 results
/ Date
/ Name
Jun 27, 2025
MPipeMoE: Memory Efficient MoE for Pre-trained Models with Adaptive Pipeline Parallelism
Apr 16, 2022
BLISS: Robust Sequence-to-Sequence Learning via Self-Supervised Input Representation
Dec 19, 2025
Enabling Disaggregated Multi-Stage MLLM Inference via GPU-Internal Scheduling and Resource Sharing
Dec 11, 2025
REMISVFU: Vertical Federated Unlearning via Representation Misdirection for Intermediate Output Feature
Sep 23, 2025
Metadata-Guided Adaptable Frequency Scaling across Heterogeneous Applications and Devices
Oct 15, 2023
Diversifying the Mixture-of-Experts Representation for Language Models with Orthogonal Optimizer
Jun 27, 2025
MCFuser: High-Performance and Rapid Fusion of Memory-Bound Compute-Intensive Operators