arXiv:2607. 19539v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) architectures increase model capacity without proportionally increasing computation cost and have become a key building block for scaling large language models (LLMs) to trillion-parameter regimes.
arXiv:2609.13585v1 Announce Type: cross Abstract: Communication has become a bottleneck in distributed training and inference of large models. Overlapping communication with computation at the granul...
arXiv:2606. 19025v1 Announce Type: cross Abstract: Pre-training Large Language Models (LLMs) typically demands large-scale infrastructure with tightly coupled hardware accelerators.
arXiv:2607. 01844v1 Announce Type: cross Abstract: This paper showcases a memory-efficient training stack for Mixture-of-Experts (MoE) models.
arXiv:2609.38090v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) models are a compelling architecture for scaling model capacity, making them especially attractive for deployment on resource-...
arXiv:2606. 10493v1 Announce Type: cross Abstract: Local deployment of large Mixture-of-Experts (MoE) models falls short of the service quality achieved in cloud-scale environments, even under low-concurrency workloads.
arXiv:2606. 00735v1 Announce Type: cross Abstract: In distributed Mixture-of-Experts (MoE) inference, input-dependent token routing interacts with GPU performance variability to create persistent stragglers under synchronized execution, where the slowest GPU determines layer latency.
KernelGenBench is a unified benchmark that evaluates large language models and agentic systems for generating efficient Triton kernels across diverse operator sources and hardware platforms. It covers 210 operators from PyTorch ATen, vLLM, and cuBLAS, and tests a 110‑operator subset on six different chips, consuming over 15 billion tokens in evaluation. The study finds that no single method dominates across all sources and platforms, with significant variations in correctness and performance depending on the operator source and hardware, and that agentic approaches require millions of tokens per successful operator.
arXiv:2608. 01563v1 Announce Type: new Abstract: Training and deployed inference often cross export, conversion, and platform-specific runtime boundaries.
arXiv:2607. 28633v1 Announce Type: cross Abstract: Disaggregated LLM inference creates a datacenter networking problem that no existing system solves correctly.
The paper reports an empirical scalability study of data‑parallel training for Kolmogorov‑Arnold Networks (KANs) on high‑performance computing systems. Using up to eight NVIDIA A100 GPUs across four nodes on the FinisTerrae III supercomputer, the authors evaluate strong and weak scaling, communication overhead, and model‑size scaling, finding a 74.7% parallel efficiency and a 5.97× speedup at eight GPUs. They observe non‑monotonic communication costs driven by All‑Reduce choices and inter‑node latency, and note that while the parameter‑to‑memory ratio improves with larger models, training time scales less favorably, leading to guidelines for GPU topology and model‑size selection.
arXiv:2603. 02376v2 Announce Type: replace-cross Abstract: Computation and communication in distributed LLM training and inference are traditionally optimized in isolation; expert-crafted systems such as DeepEP, FLUX, and TokenWeave show the potential of co-design but require deep systems expertise and hardware-specific tuning; CUCo is an agentic framework that automates compute-communication co-design of CUDA kernels by combining a structured design-space formalization with a correctness-first fast-path agent for reliable baselines and an evolution-driven slow-path agent for high-performance strategies, achieving up to 1.