Yu Guo, Shenghao Ye, Shuangwu Chen, Zijian Wen, Tao Zhang, Bai Qirui, Dong Jin, Yunpeng Hou, Huasen He, Jianyang, Xiaobin Tan. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Zijian Wen, Tao Zhang, Shuangwu Chen, Shenghao Ye, Yu Guo, Qirui Chen, Jingxian Shuai, Yunpeng Hou, Huasen He, Jianyang. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

The massive parameter scale of sparsely-activated Mixture-of-Experts (MoE) models necessitates distributed training with hybrid parallelism. Placing such training tasks,i.e.mapping the logical partitions of an MoE model to available physical NPUs, is challenging. Due to the bandwidth and latency discrepancies between intra- and inter- Pods, the cross-Pod communication usually becomes a bottleneck. The high dispersion of NPUs in multi-tenant clusters exacerbates this issue further. However, a paucity of studies has considered the cross-Pod model placement problem. To address this challenge, we propose a novel model placement scheme tailored for MoE model training with hybrid parallelism in multi-tenant clusters. By quantifying the cross-Pod communication overhead incurred during MoE model training, the model placement is formulated as a 0-1 integer quadratic problem, which is NP hard. Motivated by the traffic difference between different parallelism, we decompose this problem into two subproblems. To solve the subproblems, we propose a lightweight two-stage algorithm based on Best-Fit strategy and neighborhood search. Experiments under different models and network topologies show that our model placement scheme can reduce cross-Pod traffic by 35.9% and cut communication time by 18.7% compared to state-of-the-art methods.

High quality video can provide viewers with better visual and immersive experience in video streaming systems. However, traditional such systems with adaptive bitrate algorithms can only provide low-resolution videos when the network conditions deteriorate. To break the strong dependency of video transmission on network, we design and demonstrate iSAW, a super-resolution-assisted intelligent adaptive WebRTC video streaming system, by using two mobile devices and a policy server. iSAW performs a lightweight and scalable super-resolution (SR) model on the mobile devices to enhance the quality of received videos, and leverages an learning-based adaptive algorithm on the policy server to jointly adjust the transmitted video resolution and SR reconstructed video resolution according to dynamic network conditions and client computing power. The enhanced real-time video is smooth and the performance is improved obviously.