All news
Digital Twin

When Can LLM Digital Twins Reduce Human Measurement? From Behavioral Fidelity to Statistical Substitutability

US & West ·

The concept of statistical substitutability is introduced to assess whether LLM-based digital twins can replace repeated human data collection while preserving valid inference. A framework grounded in mixed-subject and prediction-powered inference evaluates four dimensions: aggregate fidelity, paired respondent-level signal, finite-sample label recovery, and stability across populations. This helps determine when twins can reduce measurement burden.

Why it matters

This is important for methodology of using digital twins in social and behavioral research, addressing the validity of substituting real data.

Relevant to DT products

Original headline
When Can LLM Digital Twins Reduce Human Measurement? From Behavioral Fidelity to Statistical Substitutability
Read the original: arXiv API — полнотекстовый поиск "digital twin"

Translation and summary are machine-generated from the source. The full article is not republished; its rights belong to the source.

Related

Digital twinUS & West

Earth System World Model for What-If Simulations: A Case Study for Terrestrial Ecosystems

An action-conditioned world-modeling framework for Earth-system emulation is proposed, allowing users to specify interventions and explore system responses. It reformulates simulator trajectories as supervision for controllable state-transition learning, going beyond passive forecasting. This is relevant for interactive scientific workflows and Earth-system digital twins.

DTCDT.ForecastarXiv API — полнотекстовый поиск "digital twin" ↗
Digital twinUS & West

Digital Twin and Extended Reality for Robot-Assisted Finishing

A cyber-physical system is presented for programming, supervision, training, and teleoperation of robot-assisted finishing of metal additive-manufactured parts. A digital twin built in Unity is delivered as a WebGL app supporting desktop and WebXR modes, exchanging robot state and pose commands with a collaborative robot via MQTT. The twin enforces kinematic and collision constraints before releasing poses and augments the virtual component with a color map of surface topography to support operator decisions.

DTEarXiv API — полнотекстовый поиск "digital twin" ↗