相关论文: A Modular and Fault-Tolerant Data Transport Framew…
Getting large language models (LLMs) to perform well on the downstream tasks requires pre-training over trillions of tokens. This typically demands a large number of powerful computational devices in addition to a stable distributed…
The High-Luminosity LHC (HL-LHC) will usher in a new era in high-energy physics. The HL-LHC experimental conditions entail an instantaneous luminosity of up to $7.5 \times 10^{34}$ cm$^{-2}$ s$^{-1}$ and up to 200 simultaneous collisions…
Hierarchical Federated Learning (HFL) has recently emerged as a promising solution for intelligent decision-making in vehicular networks, helping to address challenges such as limited communication resources, high vehicle mobility, and data…
The rapid increase in data traffic demand has overloaded existing cellular networks. Planned upgrades in the communication architecture (e.g. LTE), while helpful, are not expected to suffice to keep up with demand. As a result, extensive…
As AI evolves, collaboration among heterogeneous models helps overcome data scarcity by enabling knowledge transfer across institutions and devices. Traditional Federated Learning (FL) only supports homogeneous models, limiting…
The usage of federated learning (FL) in Vehicular Ad hoc Networks (VANET) has garnered significant interest in research due to the advantages of reducing transmission overhead and protecting user privacy by communicating local dataset…
ALICE analyses mostly deal with large datasets using the distributed Grid infrastructure. In LHC running periods 1 and 2, ALICE developed a system of analysis trains (so-called $"$LEGO trains$"$) that allowed the user to configure analysis…
This paper introduces Helix, a distributed system for high-throughput, low-latency large language model (LLM) serving in heterogeneous GPU clusters. The key idea behind Helix is to formulate inference computation of LLMs over heterogeneous…
Grant-Free (GF) access has been recognized as a promising candidate for Ultra-Reliable and Low-Latency Communications (URLLC). However, even with GF access, URLLC still may not effectively gain high reliability and millimeter-level latency,…
Recently we presented TTC, a domain-specific compiler for tensor transpositions. Despite the fact that the performance of the generated code is nearly optimal, due to its offline nature, TTC cannot be utilized in all the application codes…
In the forthcoming 6G era, extend reality (XR) has been regarded as an emerging application for ultra-reliable and low latency communications (URLLC) with new traffic characteristics and more stringent requirements. In addition to the…
A Large Ion Collider Experiment (ALICE) at the Large Hadron Collider (LHC) at CERN went through a major upgrade in which some of its subdetectors were replaced with new ones, while others are equipped with new electronics. The aim of the…
This paper presents an energy-efficient transmission framework for federated learning (FL) in industrial Internet of Things (IIoT) environments with strict latency and energy constraints. Machinery subnetworks (SNs) collaboratively train a…
Physics collisions at 13 TeV are expected at the LHC with an average of 40-50 proton-proton collisions per bunch crossing under nominal conditions. Tracking at trigger level is an essential tool to control the rate in high-pileup conditions…
This paper presents a novel holistic deep learning framework that simultaneously addresses the challenges of vulnerability to input perturbations, overparametrization, and performance instability from different train-validation splits. The…
Federated Learning (FL) is a decentralized machine learning paradigm where models are trained on distributed devices and are aggregated at a central server. Existing FL frameworks assume simple two-tier network topologies where end devices…
An important ingredient of the future 5G systems will be Ultra-Reliable Low-Latency Communication (URLLC). A way to offer URLLC without intervention in the baseband/PHY layer design is to use interface diversity and integrate multiple…
Dynamic programming (DP) based algorithms are essential yet compute-intensive parts of numerous bioinformatics pipelines, which typically involve populating a 2-D scoring matrix based on a recursive formula, optionally followed by a…
ALICE (A Large Ion Collider Experiment) is a detector designed to exploit the physics potential of nucleus-nucleus interactions at the LHC. Being a general purpose experiment, it will allow a comprehensive study of hadrons, electrons, muons…
Studies of heavy-ion collisions at the LHC will benefit from an array of qualitatively new probes not readily available at lower collision energies. These include fully formed jets at ET > 50 GeV, Z0's and abundantly produced heavy flavors.…