8 papers
MoE-Prefill: Zero Redundancy Overheads in MoE Prefill Serving
Zhaoyuan Su, Olatunji Ruwase, Karthik Ganesan +5
Production LLM workloads increasingly serve discriminative tasks, such as classification, recommendation, and verification, whose answers are read from the logits of a single prefi…
Shift Parallelism: Low-Latency, High-Throughput LLM Inference for Dynamic Workloads
Mert Hidayetoglu, Aurick Qiao, Michael Wyatt +3
Efficient parallelism is necessary for achieving low-latency, high-throughput inference with large language models (LLMs). Tensor parallelism (TP) is the state-of-the-art method fo…
OWL: Overcoming Window Length-Dependence in Speculative Decoding for Long-Context Inputs
Jaeseong Lee, seung-won hwang, Aurick Qiao +3
Speculative decoding promises faster inference for large language models (LLMs), yet existing methods fail to generalize to real-world settings. Benchmarks typically assume short c…
SuffixDecoding: Extreme Speculative Decoding for Emerging AI Applications
Gabriele Oliaro, Zhihao Jia, Daniel Campos +1
Speculative decoding is widely adopted to reduce latency in large language model (LLM) inference by leveraging smaller draft models capable of handling diverse user tasks. However,…
STUN: Structured-Then-Unstructured Pruning for Scalable MoE Pruning
Jaeseong Lee, seung-won hwang, Aurick Qiao +3
Mixture-of-experts (MoEs) have been adopted for reducing inference costs by sparsely activating experts in Large language models (LLMs). Despite this reduction, the massive number…
Arctic Inference with Shift Parallelism: Fast and Efficient Open Source Inference System for Enterprise AI
Samyam Rajbhandari, Mert Hidayetoglu, Aurick Qiao +5
Inference is now the dominant AI workload, yet existing systems force trade-offs between latency, throughput, and cost. Arctic Inference, an open-source vLLM plugin from Snowflake…