arXiv:2509. 23722v2 Announce Type: replace-cross Abstract: Pipeline parallelism is widely used to train large language models (LLMs).
arXiv:2609.39350v1 Announce Type: cross Abstract: As model sizes continue to scale, distributed training has become inevitable. Automatic parallelization techniques can derive efficient training para...
As model sizes continue to scale, distributed training has become inevitable. Automatic parallelization techniques can derive efficient training parallelism strategies at low cost while achieving supe...
arXiv:2509.21275v5 Announce Type: replace-cross Abstract: Long context training is crucial for extending LLM context windows. Existing schemes, such as sequence parallelism, incur substantial communi...
arXiv:2602. 24044v2 Announce Type: replace-cross Abstract: Large Language Model (LLM) adapters enable low-cost model specialization, but introduce complex caching and scheduling challenges in distributed serving systems where hundreds of adapters must be hosted concurrently.
arXiv:2608. 15762v1 Announce Type: cross Abstract: Container-granularity scheduling leaves abundant short-lived idle slices within containers unexploited.
arXiv:2605.21603v1 Announce Type: cross Abstract: Intra-device parallelism addresses resource under-utilization in ML inference and training by overlapping the execution of operators with different r...
arXiv:2512. 10236v2 Announce Type: replace-cross Abstract: Modern ML workloads demand distributing training and inference across multiple GPUs.
arXiv:2601. 20408v2 Announce Type: replace-cross Abstract: Enterprise LLM deployment faces a critical scalability challenge: organizations must optimize models systematically to scale AI initiatives within constrained compute budgets, yet the specialized expertise required for manual optimization remains a niche and scarce skillset.
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.
PipeLive introduces a method for live, in‑place reconfiguration of pipeline parallelism in large language model serving. By redesigning the KV cache layout and extending PageAttention, it enables dynamic resizing of the cache without interrupting inference. The system also uses an incremental KV patching mechanism to keep KV states consistent during reconfiguration, achieving significant reductions in reconfiguration time and improvements in latency metrics.
arXiv:2604. 26963v2 Announce Type: replace-cross Abstract: Large language models (LLMs) are increasingly deployed as the execution core of autonomous agents rather than as standalone text generators.
Para-Pipe is a hierarchical mapping framework that integrates intra- and inter-stage operator parallelism within a pipelined architecture for machine‑learning computational graphs on heterogeneous System‑on‑Chip (SoC) platforms. By selectively fine‑tuning parallelism levels across pipeline stages, it navigates the trade‑off between throughput and latency, reducing inter‑processor communication overhead and improving energy efficiency. Evaluation on Amlogic and Black Sesame SoCs shows multiple Pareto‑optimal configurations, with throughput‑optimized setups achieving up to 11.0% better energy efficiency than purely pipelined strategies and 23.3% better than non‑pipelined parallel execution.