Research Papers
Research Papers news — Friday 23 October 2026
25 stories from 2 sources, every one in English. Each links to its publisher.
-
Redundancy Meets Synergy: Dependency-aware Expert Selection for MoE via Submodular Optimization
arXiv cs.DC (Distributed)
arXiv:2610.00558v3 Announce Type: replace-cross Abstract: While Mixture-of-Experts (MoE) models effectively scale model capacity through sparse activation, their deployment is often bottlenecked by prohibitive memory requirements…
-
Parallel Sparse Outputs for Tensor Compilers
arXiv cs.DC (Distributed)
arXiv:2610.19308v2 Announce Type: replace Abstract: Sparse tensor compilers are commonly used high-performance tools that optimize tensor algebra programs over notoriously complex sparse data structures. Despite the ever-increasing size…
-
LatentStore: An Efficient Latent-First Storage System for AI-Generated Images
arXiv cs.DC (Distributed)
arXiv:2605.19385v3 Announce Type: replace Abstract: The explosive growth of AI-generated images has created a sustainability challenge for storage infrastructure. Platforms like Midjourney and Adobe Firefly already host billions of…
-
Fast Tendermint: Speeding Up a Foundational Consensus Protocol
arXiv cs.DC (Distributed)
arXiv:2608.13434v2 Announce Type: replace Abstract: Tendermint is among the most widely studied and deployed Byzantine fault-tolerant (BFT) consensus protocols, owing in part to its native leader-rotation mechanism that subsumes complex…
-
Multi-center Medical Data Mining with FL-Net - A One-stop Shop for Federated Learning
arXiv cs.DC (Distributed)
arXiv:2609.20650v2 Announce Type: replace-cross Abstract: Federated learning enables collaborative training without sharing patient-level data, but most studies remain simulations. Based on five requirements derived from the literature, we…
-
Temporal Liquidity Crossing for Blockchains: An EIP-1559 Extension for Admission Below the Base Fee
arXiv cs.DC (Distributed)
arXiv:2610.30595v1 Announce Type: cross Abstract: Transactions differ in how their value changes with execution time and relative position. Ethereum's EIP-1559 lets them compete through fee fields, but a fee bid gives no enforceable claim…
-
Evaluating the Applicability of LLM Inference Optimizations on Apple Silicon
arXiv cs.DC (Distributed)
arXiv:2610.31479v1 Announce Type: cross Abstract: Apple Silicon is among the most widely used platforms for local LLM inference, yet nearly all inference optimizations are designed and validated on server-class NVIDIA GPUs under batched…
-
Specter: Fast MoE Inference via Quantization-Enabled Multi-Step Speculative Offloading
arXiv cs.DC (Distributed)
arXiv:2511.14102v3 Announce Type: replace-cross Abstract: Mixture-of-experts (MoE) models often exceed a single accelerator's memory capacity. Offloading experts to host memory creates a severe PCIe I/O bottleneck because data-dependent…
-
Repair Changes Which Layers to Remove: Exact Optima for Depth Pruning of Language Models
arXiv cs.DC (Distributed)
arXiv:2610.30552v1 Announce Type: new Abstract: Depth pruning removes entire transformer layers to make large language models faster, and recent methods recover quality by replacing the removed layers with a training-free linear map. Which…
-
Low-Latency MoE Inference in Space Data Centers: A Wireless-Power-Transfer Perspective
arXiv cs.DC (Distributed)
arXiv:2610.30923v1 Announce Type: new Abstract: Space data centers (SpaceDCs) are emerging as platforms for large language models (LLMs), supported by continuous solar-energy harvesting and high-speed interconnections. The…
-
XIR: A Framework for Interoperability across Cross-Chain Protocols Based on a Verifiable Intermediate Representation
arXiv cs.DC (Distributed)
arXiv:2609.20010v5 Announce Type: replace-cross Abstract: Cross-chain protocols enable applications to exchange messages across blockchains. Under point-to-point configurations, communication depends on a direct connection between the…
-
Pipelining vs. Combining in Aggregation Tasks
arXiv cs.DC (Distributed)
arXiv:2610.31383v1 Announce Type: new Abstract: We study the aggregation task (convergecast) in networks in the case that the values to be aggregated do not fit in a single message. We assume that every node starts with an input value, and…
-
uGDS: A User-Space Development Kit for High-Performance GPU Direct Storage
arXiv cs.DC (Distributed)
arXiv:2610.30868v1 Announce Type: new Abstract: Data-intensive GPU applications increasingly rely on NVMe SSDs to extend GPU memory. NVIDIA GPU Direct Storage (GDS) enables direct SSD-GPU transfers via PCIe peer-to-peer DMA, eliminating…
-
On the Limits of Causal Observation in Shared-Memory Systems
arXiv cs.DC (Distributed)
arXiv:2606.14093v2 Announce Type: replace Abstract: Reasoning about concurrent system correctness requires detecting real-time operation precedence (<_H). In asynchronous shared-memory systems, real-time precedence and information-flow…
-
ATTUNE: Agentic Throughput Tuning for LLM Training
arXiv cs.DC (Distributed)
arXiv:2610.17528v2 Announce Type: replace Abstract: The throughput of an LLM training job depends on its recipe: how the model is split across GPUs, how memory is traded for compute, and which kernels and communication overlaps it uses…
-
A mechanism for collective foraging: percolating recruitment cascades coupled to a maximum-entropy field
arXiv cs.DC (Distributed)
arXiv:2610.24865v3 Announce Type: replace-cross Abstract: Foraging recruitment is a contagion: an agent that finds a resource triggers a linear-threshold cascade, the model built for rumour and adoption spread on social networks, now…
-
ASA: A Simulation Environment for Evaluating Military Operational Scenarios
arXiv cs.DC (Distributed)
arXiv:2207.12084v2 Announce Type: replace Abstract: The Aerospace Simulation Environment (Ambiente de Simula\c{c}\~ao Aeroespacial -- ASA in Portuguese) is a custom-made object-oriented simulation framework developed mainly in C++ that…
-
MCP vs CLI for AI Agents: A Comparative Analysis of Tool Integration Strategies for Autonomous Systems
arXiv cs.DC (Distributed)
arXiv:2610.30157v1 Announce Type: cross Abstract: AI agents require interfaces to interact with external systems such as files, APIs, databases, and cloud services. Two competing paradigms have emerged: CLI (Command Line Interface), where…
-
Cambium: Participant-Built Commitment Roots for Shared Digital Infrastructure
arXiv cs.DC (Distributed)
arXiv:2610.30409v1 Announce Type: new Abstract: Commitment services, such as timestamping, transparency logs and provenance registries, let a party prove that it bound itself to a record without publishing the record. Today an operator…
-
Evolving Federated Density Peak Clustering with Shuffler-Based Anonymization
arXiv cs.DC (Distributed)
arXiv:2610.31211v1 Announce Type: cross Abstract: In this paper, we present a novel one-shot federated clustering approach based on evolving fuzzy clustering and shuffler-based anonymization. In federated clustering, each round reveals…
-
Almost tight distributed approximation for domination in sparse graph classes
arXiv cs.DC (Distributed)
arXiv:2610.31203v1 Announce Type: new Abstract: We study distributed approximation of Minimum Dominating Set in a very general class of sparse graphs, extending near-optimal approximation results from the minor-free setting to bounded…
-
Know When to Escalate: Critic-Guided Adaptive Execution for Cloud-Local Embodied Agents
arXiv cs.DC (Distributed)
arXiv:2610.30587v1 Announce Type: cross Abstract: Recent advances in foundation-model agents have enabled increasingly capable embodied systems for long-horizon planning and interaction. However, embodied agents require fast, closed-loop…
-
Procrustes-Bering-Sheaft: Automated Model Discovery and Virtual Crash Tests for Microservices
arXiv cs.DC (Distributed)
arXiv:2610.31444v1 Announce Type: cross Abstract: Procrustes-Bering-Sheaft automates lightweight model discovery and virtual crash tests for microservices. Each test estimates operation availability by Monte Carlo sampling of service and…
-
Hard and Soft Real-Time DNN Inference Scheduling on Heterogeneous Multi-Accelerator Systems via Reinforcement Learning
arXiv cs.DC (Distributed)
arXiv:2610.29503v2 Announce Type: replace-cross Abstract: We present a scheduling framework for heterogeneous multi-accelerator systems executing both hard real-time and soft real-time QoS-aware DNN inference workloads. The framework…
-
Putting conversational AI to the test in primary care
Nature Technology