CBA: Communication-Bound-Aware Cross-Domain Resource Assignment for Pipeline-Parallel Distributed LLM Training in Dynamic Multi-DC Optical Networks

2025-12-23 · Dianxuan Fu, Xiaomin Liu, Yihao Zhang, Shikui Shen, Weisheng Hu, Qunbi Zhuge arxiv

We propose a communication-bound-aware cross-domain resource assignment framework for pipeline-parallel distributed training over multi-datacenter optical networks, which lowers iteration time by 31.25% and reduces 13.20% blocking requests compared to baselines.

📄 PDF Abstract BibTeX arXiv:2512.20080

Code (0)

등록된 구현이 없습니다.

Social-Mobility-Aware Joint Communication and Computation Resource Management in NOMA-Enabled Vehicular Networks

2023-07-08 · Tong Xue, Haixia Zhang, Hui Ding, Dongfeng Yuan

The existing computation and communication (2C) optimization schemes for vehicular edge computing (VEC) networks mainly focus on the physical domain without considering the influence from the social domain. This may grea…

Edge-computing Management

DCMAC: Demand-aware Customized Multi-Agent Communication via Upper Bound Training

2024-09-11 · Dongkun Huo, Huateng Zhang, Yixue Hao, Yuanlin Ye 외

Efficient communication can enhance the overall performance of collaborative multi-agent reinforcement learning. A common approach is to share observations through full communication, leading to significant communication…

Multi-agent Reinforcement Learning

Resource allocation for text semantic communications

2022-01-16 · Lei Yan, Zhijin Qin, Rui Zhang, Yongzhao Li 외

Semantic communications have shown its great potential to improve the transmission reliability, especially in the low signal-to-noise regime. However, resource allocation for semantic communications still remains unexplo…

Semantic Communication

Domain-Adaptive Transformer for Data-Efficient Glioma Segmentation in Sub-Saharan MRI

2025-11-04 · Ilerioluwakiiye Abolade, Aniekan Udo, Augustine Ojo, Abdulbasit Oyetunji 외 arxiv

Glioma segmentation is critical for diagnosis and treatment planning, yet remains challenging in Sub-Saharan Africa due to limited MRI infrastructure and heterogeneous acquisition protocols that induce severe domain shif…

Domain Adaptation

RABC-Net: Reliability-Aware Annotation-Free Skin Lesion Segmentation for Low-Resource Dermoscopy

2026-04-07 · Yujie Yao, Yuhaohang He, Junjie Huang, Zhou Liu 외 arxiv

Pixel-level annotation is costly in low-resource dermoscopy. We present RABC-Net, a reliability-aware annotation-free segmentation system that combines pseudo-label reliability learning, restricted target-domain adaptati…

Skin Lesion Segmentation Domain Adaptation