2 papers
cs.CV2026
Accelerating Diffusion via Hybrid Data-Pipeline Parallelism Based on Conditional Guidance Scheduling
Euisoo Jung, Byunghyun Kim, Hyunjin Kim +2
Diffusion models have achieved remarkable progress in high-fidelity image, video, and audio generation, yet inference remains computationally expensive. Nevertheless, current diffu…
cs.DC2026
TimelyFreeze: Adaptive Parameter Freezing Mechanism for Pipeline Parallelism
Seonghye Cho, Jaemin Han, Hyunjin Kim +2
Pipeline parallelism enables training models that exceed single-device memory, but practical throughput remains limited by pipeline bubbles. Although parameter freezing can improve…