AX·NEWS
Sign in

Research Papers

Research Papers news — Friday 20 November 2026

14 stories from 2 sources, every one in English. Each links to its publisher.

Open in the live feedRSS feedOn a TV

  1. HEL: A Heterogeneous Extensible Layer for GPU-Resident Runtime Orchestration
    USarXiv cs.DC (Distributed)

    arXiv:2610.24830v2 Announce Type: replace Abstract: We describe a GPU-resident runtime system, HEL, built on stateful services that persist throughout the lifespan of a user application. A HEL programmer expresses a workload as an…

  2. Beyond 4: Improved Competitive Traveling Repairperson
    USarXiv cs.DC (Distributed)

    arXiv:2611.21943v1 Announce Type: cross Abstract: With $k\geq 1$ repairpersons starting initially from a depot, we present a 3.877-competitive deterministic algorithm for the fundamental online traveling repairperson problem on an…

  3. Cooling Noisy Population Protocols -- Vanishing-Gain Averaging and Localisation under Persistent Errors
    USarXiv cs.DC (Distributed)

    arXiv:2611.26266v1 Announce Type: new Abstract: We study noise-tolerant coordination in anonymous population protocols whose agents interact in uniformly random pairs, maintain real-valued labels, and have no global clock. Each agent cools…

  4. AscMLA: Unlocking FP4 Efficiency for Long-Context MLA Decoding through Asymmetric Scaling
    USarXiv cs.DC (Distributed)

    arXiv:2611.26106v1 Announce Type: new Abstract: Long-context decoding repeatedly reads a growing KV cache. FP4 reduces cache storage and traffic but needs block scaling to handle its limited dynamic range. Applying FP4 to multi-head latent…

  5. Triple Overlap in Pipelined Krylov Methods on Multi-GPU Systems
    USarXiv cs.DC (Distributed)

    arXiv:2610.25058v2 Announce Type: replace Abstract: We introduce a novel triple-overlap implementation of the pipelined preconditioned conjugate gradient (Pipe-PCG) and pipelined stabilized biconjugate gradient (Pipe-BiCGStab) solvers…

  6. Towards Scale-Up Networking: Challenges and Open Questions
    USarXiv cs.DC (Distributed)

    arXiv:2611.26156v1 Announce Type: cross Abstract: The rapid growth of AI models and the widespread use of distributed training and inference have made inter-accelerator communication a critical system bottleneck. Scale-up fabrics address…

  7. Berth: SLO-Preserving GPU Sharing Without New Preemption
    USarXiv cs.DC (Distributed)

    arXiv:2611.25091v1 Announce Type: new Abstract: Inference services keep GPUs reserved for their worst-case load, and most of that capacity sits idle. Filling it with best-effort work such as training is the obvious remedy, and latency is…

  8. Thinflow: Organizing HPC Benchmarking Campaigns as Rerunnable Directory Trees
    USarXiv cs.DC (Distributed)

    arXiv:2611.26526v1 Announce Type: new Abstract: A benchmarking campaign provides the measurements behind an HPC performance study: code variants are built, swept over core or node counts, and measured repeatedly. Moving such a campaign to…

  9. Properties of Concurrent Algorithms that Allow Asynchrony
    USarXiv cs.DC (Distributed)

    arXiv:2302.14834v4 Announce Type: replace-cross Abstract: In this paper, we investigate the conditions under which a concurrent system is able to guarantee convergence deterministically even without synchronization. In general, an…

  10. Accelerating Mixture of Block Attention via Dynamic Routing, KV-Centric Execution and Runtime Pipelining
    USarXiv cs.DC (Distributed)

    arXiv:2611.26268v1 Announce Type: new Abstract: Mixture of Block Attention (MoBA) reduces long-context computation by selecting key-value (KV) blocks for each query. FlashMoBA implements MoBA efficiently but retains a fixed Top-k budget…

  11. StarMPI-SVD: Distributed-Memory Parallel Star-M Tensor SVD
    USarXiv cs.DC (Distributed)

    arXiv:2611.26451v1 Announce Type: new Abstract: The star-M tensor algebra generalizes the optimal compression guarantees of the matrix singular value decomposition (SVD) to higher-order tensors. The t-SVDM-I decomposition is computed using…

  12. Twindex: Index Cache Offloading with Coarse-to-Fine Selection for Long-Context Sparse-Attention Serving
    USarXiv cs.DC (Distributed)

    arXiv:2611.25078v1 Announce Type: new Abstract: In long-context decoding, GPU memory (HBM) caps how many requests a GPU decodes at once. Sparse-attention models such as Qwen3.8-Flash-Next, DeepSeek V4.1-Flash and GLM-5.2 add an indexer: at…

  13. Live Migration of Gated DeltaNet Requests: Exact Catch-Up and the Cost of Log Backlog
    USarXiv cs.DC (Distributed)

    arXiv:2611.26153v1 Announce Type: new Abstract: Live migration of a streaming Gated DeltaNet (GDN) request requires recovering recurrent state at the correct token boundary on a worker with the same model weights. We adapt snapshots…

  14. How AI is affecting science careers
    GBNature Technology