Digital Twins

Follow this topic Ask about this topic

Digital twins are computational counterparts of physical assets, people, or broader systems, updated with observations to represent their changing state and test possible actions. Approaches range from physics-based simulation and data assimilation to machine-learning surrogates, hybrid models, generative agents, and reinforcement learning for control. Applications span manufacturing, infrastructure, transportation, healthcare, energy, and social systems, where twins can support forecasting, maintenance, safety testing, and personalized decisions. A central challenge is ensuring that a twin remains calibrated and trustworthy as conditions change, while balancing fidelity, uncertainty, computation, privacy, and validation against real-world outcomes.

Latest papers

253 papers36 papers in the last four weeks, up 16% on the four weeks before. 0.2% of all new papers.

Data Driven Optimization of GPU efficiency for Distributed LLM-Adapter Serving
Feb 27, 2026Ferran Agullo, Joan Oliveras, Chen Wang +5LLM Inference Efficiency LLM Serving
Barcelona Supercomputing Center (BSC), Barcelona, Spain · Universitat Politècnica de Catalunya - BarcelonaTech (UPC), Barcelona, Spain · IBM Research, New York, USA
Multi-Condition Digital Twin Calibration for Axial Piston Pumps : Compound Fault Simulation
Feb 27, 2026Chang Dong, Jianfeng Tao, Chengliang LiuDigital Twins Fault Diagnosis
State Key Laboratory of Mechanical System and Vibration, Shanghai Jiao Tong University, Shanghai 200240, China
Simple Models, Real Swimming: Digital Twins for Tendon-Driven Underwater Robots
Feb 26, 2026Mike Y. Michelis, Nana Obayashi, Josie Hughes +1Underwater Robotics Robotics Simulation
Soft Robotics Lab and ETH AI Center, ETH Zurich, Switzerland · Prema Lab, New York University, USA · CREATE Lab, EPFL, Switzerland
Data-Driven Supervision of a Thermal-Hydraulic Process Towards a Physics-Based Digital Twin
Feb 25, 2026Osimone Imhogiemhe, Yoann Jus, Hubert Lejeune +1Fault Detection Process Supervision
LS2N, LS2N - équipe SIMS, Nantes Univ - ECN · Nantes Univ., Centrale Nantes LS2N, CNRS UMR 6004 F-44000 Nantes, France · Cetim +1
Personalized Prediction of Perceived Message Effectiveness Using Large Language Model Based Digital Twins
Feb 23, 2026Jasmin Han, Janardan Devkota, Joseph Waring +8Digital Twins Personalized Language Models
Department of Mental Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA · Department of Health Sciences, Towson University, Towson, USA · Addiction Research Group, University of East Anglia, Norwich, UK +8
Artificial Intelligence for Modeling & Simulation in Digital Twins
Feb 22, 2026Philipp Zech, Istvan DavidDigital Twins
University of Innsbruck, Austria · McSCert, McMaster University, Canada
EDU-MATRIX: A Society-Centric Generative Cognitive Digital Twin Architecture for Secondary Education
Feb 21, 2026Wenjing Zhai, Jianbin Zhang, Tao LiuDigital Twins Human Behavior Simulation
The High School Affiliated to Beijing Normal University, Beijing, China · Department of Electronic and Communication Engineering, North China Electric Power University, Hebei, China
Federated Learning-Assisted Optimization of Mobile Transmission with Digital Twins
Feb 20, 2026Mohammad Heydari, Terence D. Todd, Dongmei Zhao +1Privacy-Preserving Federated Learning Wireless Resource Allocation
Department of Electrical and Computer Engineering, McMaster University, CANADA · Department of Computing and Software, McMaster University, CANADA
An Industrial Dataset for Scene Acquisitions and Functional Schematics Alignment
Feb 17, 2026Flavien Armangeon, Thibaud Ehret, Enric Meinhardt-Llopis +4Cross-Modal Alignment Digital Twins
Universit´e Paris-Saclay, ENS Paris-Saclay, CNRS, Centre Borelli, 91190, Gif-sur-Yvette, France · EDF R&D (Electricit´e de France), France · AMIAD, Pˆole Recherche, France +1
Digital Twin-Driven Communication-Efficient Federated Anomaly Detection for Industrial IoT
Jan 5, 2026Mohammed Ayalew Belay, Adil Rasheed, Pierluigi Salvo RossiCommunication-Efficient Distributed Training Industrial Anomaly Detection
Department of Electronic Systems, Norwegian University of Science and Technology, 7034 Trondheim, Norway
Psychometric Comparability of LLM-Based Digital Twins
Dec 22, 2025Yufei Zhang, Zhihao MaLLM Evaluation Personality Modeling in Language Models
Computational Communication Collaboratory, School of Journalism and Communication, Nanjing University, Nanjing 210023, China
Adaptive digital twins for predictive decision-making: Online Bayesian learning of transition dynamics
Dec 15, 2025Eugenio Varetti, Matteo Torzoni, Marco Tezzele +1System Identification Digital Twins
MOX – Dipartimento di Matematica, Politecnico di Milano, Milan, 20133, Italy · Dipartimento di Ingegneria Civile e Ambientale, Politecnico di Milano, Milan, 20133, Italy · Mathematics Department, Emory University, Atlanta, 30322, GA, United States
On the Use of AI-Driven Immersive Digital Technologies for Designing and Operating UAVs
Jul 23, 2024Yousef Emami, Mohammadhossein Homaei, Miguel Gutierrez Gaitan +2Aerial Robotics Digital Twins

Feb 27, 2026J/K move · Enter open · S save

Data Driven Optimization of GPU efficiency for Distributed LLM-Adapter Serving

Ferran Agullo, Joan Oliveras, Chen Wang, Alberto Gutierrez-Torre+4

Barcelona Supercomputing Center (BSC), Barcelona, Spain · Universitat Politècnica de Catalunya - BarcelonaTech (UPC), Barcelona, Spain · IBM Research, New York, USA

LLM Inference Efficiency LLM Serving Digital Twins

Open paper AskPDF arXiv

Large Language Model (LLM) adapters enable low-cost model specialization, but introduce complex caching and scheduling challenges in distributed serving systems where hundreds of adapters must be hosted concurrently. While prior work has largely focused on latency and throughput optimization, minimizing GPU resource requirements through near-peak utilization remains largely underexplored. This paper presents a data-driven pipeline that, for a given workload, computes an adapter placement that serves the workload with the minimum number of GPUs while avoiding request starvation and GPU memory errors. To that end, the approach identifies the maximum feasible throughput attainable on each GPU by leveraging accurate performance predictions learned from real serving behavior. The proposed pipeline integrates three components: (i) a Digital Twin (DT) tailored to LLM-adapter serving, (ii) a distilled machine learning (ML) model trained on DT-generated data, and (iii) a greedy placement algorithm that exploits ML-based performance estimates to maximize GPU efficiency. The DT emulates real system dynamics with high fidelity, achieving below 5% throughput estimation error while executing up to 90x faster than full LLM benchmarking across both predictable and unpredictable workloads. The learned ML models further accelerate performance estimation with marginal accuracy degradation, enabling scalable optimization. Experimental results demonstrate that the pipeline substantially improves GPU efficiency, reducing the number of GPUs required to sustain target workloads by 60% on average across the evaluated scenarios. Beyond GPU efficiency, the pipeline can be adapted to alternative objectives, such as latency minimization, highlighting its versatility for future large-scale LLM serving infrastructures.