Discriminative models, such as Convolutional Neural Networks and Vision Transformers, have driven significant progress in medical image segmentation, their limited generalization ability across diverse datasets remains a key challenge. Consequently, generative models have been increasingly explored for such image-to-image tasks due to their capacity to produce flexible and diverse outputs. Among these, methods based on diffusion models have shown considerable promise. A critical drawback, however, is that existing approaches often fail to adapt these models natively for segmentation, leading to compromised performance. To address this limitation, we introduce MSLD, a novel framework for medical semantic segmentation based on latent diffusion models. Our approach recasts segmentation as a conditional generation problem, where the model is fine-tuned to generate a segmentation mask conditioned on the input image. Extensive experiments on multiple datasets demonstrate that our proposed method achieves high-quality medical image segmentation.

Mixture-of-Expert (MoE) models outperform conventional models by selectively activating different subnets, named experts, on a per-token basis. This gated computation generates dynamic communications that cannot be determined beforehand, challenging the existing GPU interconnects that remain static during distributed training. In this paper, we advocate for a first-of-its-kind system, called MixNet, that unlocks topology reconfiguration during distributed MoE training. Towards this vision, we first perform a production measurement study and show that the MoE dynamic communication pattern has strong locality, alleviating the need for global reconfiguration. Based on this, we design and implement a regionally reconfigurable high-bandwidth domain that augments existing electrical interconnects using optical circuit switching (OCS), achieving scalability while maintaining rapid adaptability. We build a fully functional MixNet prototype with commodity hardware and a customized collective communication runtime. Our prototype trains state-of-the-art MoE models with in-training topology reconfiguration across 32 A100 GPUs. Large-scale packet-level simulations show that MixNet achieves performance comparable to a non-blocking fat-tree fabric while boosting the networking cost efficiency (e.g., performance per dollar) of four representative MoE models by 1.2×–1.5× and 1.9×–2.3× at 100 Gbps and 400 Gbps link bandwidths, respectively.