NSDI 2026

Meta Info

Homepage: https://www.usenix.org/conference/nsdi26

Acceptance Rate

• Spring: 24.2% (= 50 / 207)

Papers

LLM

• LLM Training

  • Attack of the Bubbles: Straggler-Resilient Pipeline Parallelism for Large Model Training [arXiv]

    • HKUST & Alibaba

• LLM Inference

  • Fast Distributed Inference Serving for Large Language Models [arXiv]

    • PKU

  • HydraServe: Minimizing Cold Start Latency for Serverless LLM Serving in Public Clouds [arXiv]

    • PKU & Alibaba Cloud

• LLM Storage

  • ZipLLM: Efficient LLM Storage via Model-Aware Synergistic Data Deduplication and Compression [arXiv] [Homepage]

    • UVA & Harvard

Acronyms

• LLM: Large Language Model