Research Papers
Research Papers news — Thursday 05 November 2026
23 stories from 3 sources, every one in English. Each links to its publisher.
-
Quantum gases below 10 picokelvin in space
Science Magazine Tech
Science, Ahead of Print.
-
TokenScale: Timely and Accurate Autoscaling for Disaggregated LLM Serving with Token Velocity
arXiv cs.DC (Distributed)
arXiv:2512.03416v2 Announce Type: replace Abstract: The architectural shift to prefill/decode (PD) disaggregation in LLM serving improves resource utilization but struggles with the bursty nature of modern workloads. Existing autoscaling…
-
Distributed Consensus, Optimization, and Learning under Communication Constraints
arXiv cs.DC (Distributed)
arXiv:2611.06524v1 Announce Type: cross Abstract: Many modern engineering applications require distributed problem-solving in which multiple nodes in a network cooperate to achieve a common optimization objective. Numerous distributed…
-
Shallow Encoding Emits Deep Memories
arXiv cs.DC (Distributed)
arXiv:2611.06084v1 Announce Type: cross Abstract: Hybrid language models interleave a few softmax-attention layers with many linear-attention or state-space layers, so every layer keeps a memory of earlier tokens: a key-value cache or a…
-
RESHPart: Feedback-Reshaped Spectral Embedding for Hypergraph Partitioning
arXiv cs.DC (Distributed)
arXiv:2611.05729v1 Announce Type: cross Abstract: With scalable Laplacian solvers now available, the next frontier is to exploit them more effectively for high-quality hypergraph partitioning. In this work, we revisit spectral hypergraph…
-
AtomicFaaS: Offloading Side-Effect Compensation to the Serverless Platform
arXiv cs.DC (Distributed)
arXiv:2611.06601v1 Announce Type: new Abstract: Function-as-a-Service (FaaS) simplifies distributed application development by offloading operational concerns to the platform, but developers remain responsible for recovering from failures…
-
Optimal Subcubic Byzantine Agreement with Plain PKI
arXiv cs.DC (Distributed)
arXiv:2611.06479v1 Announce Type: new Abstract: We consider Byzantine Agreement (BA), where $n$ parties run a distributed protocol to agree on a common value, despite $t$ parties being byzantine. In the synchronous setting, a choice must…
-
Weaver: Unlocking Blockchain Transaction Parallelism through State-Decoupled Symbolic Execution
arXiv cs.DC (Distributed)
arXiv:2611.06462v1 Announce Type: new Abstract: Transaction execution remains a major performance bottleneck in blockchains. Existing parallel execution engines are still blocked by slow state accesses on each transaction's critical path…
-
Performance Portability of Out-of-Kernel Optimizations on GPUs: A Molecular Docking Case Study
arXiv cs.DC (Distributed)
arXiv:2611.06207v1 Announce Type: new Abstract: Achieving performance portability is challenging for irregular scientific applications built as repeated multi-kernel GPU pipelines, because throughput depends on out-of-kernel decisions as…
-
Efficient Stable Word-RAM Simulation by Population Protocols
arXiv cs.DC (Distributed)
arXiv:2611.06178v1 Announce Type: new Abstract: Population protocols model computation by indistinguishable agents interacting in uniformly random pairs, with unary inputs and outputs encoded by state counts. Efficiency requires…
-
CARE-FL: Federated Fine-Tuning of Multimodal LLMs in Resource-Constrained Edge Networks
arXiv cs.DC (Distributed)
arXiv:2611.06169v1 Announce Type: new Abstract: Domain-specific adaptation of Multimodal Large Language Models (MLLMs) in privacy-sensitive edge environments requires federated fine-tuning, yet client selection remains challenging due to…
-
Beyond Compression Ratio: High-Throughput KV Compression for Disaggregated LLM Serving
arXiv cs.DC (Distributed)
arXiv:2611.04382v2 Announce Type: replace Abstract: KV cache compression reduces the communication volume of disaggregated large language model serving, but its practical benefit depends not only on compression ratio and model quality, but…
-
Energy-Efficient Reinforcement Learning for Large Language Models
arXiv cs.DC (Distributed)
arXiv:2611.05748v1 Announce Type: new Abstract: Reinforcement learning for language models alternates between generating responses and training the policy. These stages place different demands on GPUs, creating opportunities to reduce…
-
HYPIC: Position-Independent Caching for Hybrid-Attention LLM Serving
arXiv cs.DC (Distributed)
arXiv:2607.01299v3 Announce Type: replace Abstract: In retrieval-augmented generation (RAG) and agentic LLM serving, prompts are assembled from multiple independent segments into long contexts, making prefill dominate the computation cost…
-
RCOMPSs: A Scalable Runtime System for R Code Execution on Manycore Systems
arXiv cs.DC (Distributed)
arXiv:2505.06896v2 Announce Type: replace Abstract: R has become a cornerstone of scientific and statistical computing due to its extensive package ecosystem, expressive syntax, and strong support for reproducible analysis. However, as…
-
Constant-Round Massively Parallel Algorithms for Approximate MSTs in Metric Spaces
arXiv cs.DC (Distributed)
arXiv:2611.06092v1 Announce Type: cross Abstract: We develop a unified algorithmic framework for computing approximate minimum spanning trees (MSTs) in metric spaces in the fully scalable Massively Parallel Computation (MPC) model. Using…
-
Easy Multi-Precision Acceleration via Vectorization
arXiv cs.DC (Distributed)
arXiv:2611.05558v1 Announce Type: cross Abstract: High level programming languages and accelerators are powerful enablers for a wide range of applications. Accelerators offer significant performance via diverse numeric precision and…
-
flash-comm: Self-Synchronizing Collective Communication for LLM Decoding
arXiv cs.DC (Distributed)
arXiv:2611.06634v1 Announce Type: new Abstract: We present flash-comm, a communication library that achieves efficient transport for LLM decoding by fusing readiness synchronization with data movement. flash-comm introduces…
-
Chiron: Cache Management Empowered by LLM Intelligence for AI Workloads
arXiv cs.DC (Distributed)
arXiv:2611.06179v1 Announce Type: new Abstract: AI clusters keep data in remote object stores and rely on local caches to stop GPUs from waiting for data. A cache performs well only when its prefetching and eviction policies match the…
-
A Controlled Experimental Evaluation of Configuration and Cache Architecture in LLM Serving Engines
arXiv cs.DC (Distributed)
arXiv:2611.05991v1 Announce Type: new Abstract: Multi-agent LLM applications repeat most of their prompt tokens across requests, and serving engines reuse the cached state of shared prefixes. vLLM and SGLang reuse it at different…
-
Runway Characters: Serving a 57B Video Diffusion Transformer for Real-Time Conversation
arXiv cs.DC (Distributed)
arXiv:2611.05642v1 Announce Type: new Abstract: Conversational video characters must generate video as fast as it plays, for one user at a time: at 24 fps, a sustained 166.7 ms per four-frame chunk at batch size one, with only a few chunks…
-
vExpert: Virtualizing Expert Storage for Adaptive Load Balancing in Distributed MoE Inference
arXiv cs.DC (Distributed)
arXiv:2611.06057v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) models have become prevailing, yet their distributed deployment suffers from severe dynamic load imbalance due to the conflict between static expert placement and…
-
Responsible use of artificial intelligence
Nature Technology