Skip to content
RESEARCH · QUANTUM MACHINE LEARNING

A small quantum computer learns to sort simulated collider debris

A new preprint tests training and evaluation on real hardware using simulated particle-collision data. It does not establish a quantum advantage.

Published Source report Sources checked
Reading tools for text size, copying, printing and your reading list
My reading list

A small quantum computer has learned to distinguish two kinds of simulated particle-collider debris, a November 12 preprint reports. The test asks whether a learning algorithm can train on noisy quantum hardware, rather than only in a simulation.[1]

The Italian team used four on the five-qubit Lagrange machine in Turin to classify particle streams associated with beauty and charm quarks. Quantum circuits ran on the chip during training and testing; a conventional processor handled the optimization.[1]

The model trained on 100 simulated jets and was tested on 1,000. Training consumed 43 minutes of quantum-processor time, excluding queuing and other elapsed-time costs. Its overall discrimination was comparable to the corresponding noiseless simulation.[1]

This is a hardware feasibility result. A separate simulator comparison with a small classical neural network did not establish quantum superiority. The dataset describes simulated LHCb collisions, not a new analysis of recorded collisions.[1]

The next question is whether the approach remains practical with larger datasets and circuits, which require more repeated quantum measurements.[1]

Keep your reading trail

Evidence & context

1 linked source

Research preprint; real quantum hardware using simulated collision data

How to read the evidence and dates

An announcement records what an organization stated. An experiment reports an observation. A simulation or resource estimate depends on its assumptions. QubitWire’s analysis connects these sources; the labels are not quality scores.

Published is the article’s first QubitWire publication. Updated marks a substantive content change. Sources checked records a review of the underlying evidence. An older paper can support a new explanation, and another press release about the same work is not independent replication.

QubitWire take

Training on a chip tests noise and resource costs; it does not by itself establish a quantum advantage.

Sources

  1. Beauty and charm jet identification with end-to-end training on a superconducting quantum processor

    Preprint v1 submitted November 12, 2026 at 10:45:24 UTC. Hardware circuit evaluations with classical CPU optimization; simulated LHCb data. First experimental disclosure date not established; no independent reproduction.

READ NEXT

Continue exploring

All Research stories
  1. Quantum software

    Small quantum-training test cuts return trips to the cloud

    Three small IBM-processor experiments show why how work is submitted can matter, not just how many measurements run.

    Published · Source report · Sources checked · About 2 min read