GF-DiT dynamically adapts parallelism during DiT serving via trajectory tasks and group-free collectives, reporting up to 6x throughput and 95% latency reduction versus static configurations.
Title resolution pending
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
verdicts
UNVERDICTED 3representative citing papers
BPL is a spinlock that batches waiting tasks by request order and prioritizes within batches to cut average delay for high-priority tasks while keeping the same worst-case bound as FIFO locks.
A two-level decoder scheduling framework reduces classical processing requirements for quantum error correction by 10-40% on fault-tolerant benchmarks by managing bursty workloads as shared resources.
citing papers explorer
-
GF-DiT: Scheduling Parallelism for Diffusion Transformer Serving
GF-DiT dynamically adapts parallelism during DiT serving via trajectory tasks and group-free collectives, reporting up to 6x throughput and 95% latency reduction versus static configurations.
-
Bounded Priority-Aware Locking for Real-Time Kernels
BPL is a spinlock that batches waiting tasks by request order and prioritizes within batches to cut average delay for high-priority tasks while keeping the same worst-case bound as FIFO locks.
-
Managing Classical Processing Requirements for Quantum Error Correction
A two-level decoder scheduling framework reduces classical processing requirements for quantum error correction by 10-40% on fault-tolerant benchmarks by managing bursty workloads as shared resources.